4566results about "Structural/machines measurement" patented technology

The invention provides a water pump residual life prediction method based on a large model, and the method comprises the following steps: S1, collecting the multi-source heterogeneous data of the operation of a water pump in real time through a vibration sensor, a temperature sensor, a pressure sensor and a monitoring unit, the temperature sensor monitors temperature gradient changes of the bearing and the sealing cavity, the pressure sensor records inlet and outlet pressure fluctuation characteristics, and the monitoring unit extracts three-phase current harmonic components of the motor; s2, carrying out lightweight preprocessing on the multi-source heterogeneous original sensing data at an edge computing node, wherein the lightweight preprocessing comprises vibration signal noise reduction processing based on wavelet transform, temperature and pressure data calibration normalization of load segments according to working conditions, and transient abnormal data flow filtering through a sliding time window; and S3, inputting the preprocessed data stream into a cloud large model platform, and analyzing the long-period dependency relationship of the vibration signals through a Transform encoder in a time sequence feature extraction module.

The invention relates to a multi-sensor fusion intelligent actuator fault self-diagnosis method and system, and belongs to the technical field of actuator fault diagnosis. The method comprises the following steps: realizing multi-sensor clock synchronization through a unified clock source, generating a synchronous time sequence for asynchronous data such as Hall, pressure and temperature by adopting an interpolation method, and intensively acquiring signals such as rotating speed, torque and vibration; based on a current sensor and temperature data, primary fault identification is carried out through a current overload protection model, and an adaptive processing mechanism is triggered; demodulating the noise-reduced vibration signal, extracting high-frequency energy and detecting the deviation degree between the fault frequency of the bearing and a base line; bearing energy consumption abnormity is detected, and a fusion weight is dynamically set by combining the deviation degree, the energy consumption coefficient and an overload result; and carrying out weighted fusion on the multi-modal features, inputting an intelligent diagnosis model to carry out fault mode identification, and finally outputting a fault type / position and triggering a processing strategy. Accurate and rapid self-diagnosis of the actuator fault is realized.

The invention relates to the technical field of fault detection, in particular to a heating and ventilation system fault positioning system and method based on big data, and the system comprises a multi-source sensing module, a disturbance feature module, a path modeling module, a frequency spectrum matching module and a fault positioning module. In the method, a real-time disturbance sequence is constructed through time window segmentation and parameter offset calculation, separation of an active response chain and an abnormal propagation path is realized through a directional joint state vector and a topological relation table, and time asynchronism of multi-device signal transmission is eliminated by adopting a dynamic time warping algorithm. In combination with a real-time parameter bidirectional verification mechanism of a frequency domain main frequency band energy mark, a valve opening degree and a pump rotating speed, the problem of path confusion in a multi-node parameter coupling scene of a traditional method is solved, the tracing efficiency of concurrent faults in a complex pipe network system is improved, the adaptability limitation of a single-dimensional threshold mechanism to equipment performance degradation is overcome, and the method is suitable for a complex pipe network system. And the error positioning probability caused by signal delay superposition is reduced.

The invention discloses a multi-modal monitoring and anomaly detection method, system and device for a belt conveying system and a medium, and belongs to the technical field of anomaly detection.The method comprises the steps that a plurality of multi-modal sensors are arranged on the whole path along the path of the belt conveying system, and original signals in the running process of the belt conveying system are collected; the original signals are preprocessed; performing environment compensation calculation to generate engineering data; mapping the engineering data to the same space coordinate system of the belt conveying system, and completing feature extraction to obtain feature vectors; multi-modal fusion analysis is carried out on the feature vectors, abnormal state identification and fault level classification are realized, and an early warning signal and a risk level are output; and executing a matched response control strategy according to the early warning signal, the risk level and the corresponding space coordinate system position. According to the method, the abnormal state can be dynamically recognized under the complex working condition, differential response control is achieved, and the recognition capability and processing precision of early-stage, multi-source and dynamic anomalies are improved.

The invention discloses a circuit breaker service life prediction system and method based on multi-source heterogeneous data fusion and dynamic weight correction, and the method comprises the steps: collecting the current, pressure, displacement, main loop current, voltage, contact temperature, environment temperature and partial discharge of a circuit breaker, and carrying out the multi-source heterogeneous data fusion algorithm, thereby achieving the prediction of the service life of the circuit breaker. Carrying out fusion analysis on the collected data; a dynamic weight correction algorithm is executed according to an analysis result, and different types of fault weights are corrected and adjusted, so that the life prediction model is more accurate; and executing a fault identification algorithm after each action of the circuit breaker, and calculating the service life of the circuit breaker according to the weight and the health index corresponding to each fault. According to the invention, the prediction accuracy of the service life of the short-circuiter is greatly improved, and the use safety of power equipment is improved.

The invention discloses an intelligent operation and maintenance method. The method comprises the steps of deploying a vibration sensor, a temperature sensor, a current sensor and a pressure sensor at key operation parts of building equipment; receiving an original data flow of each sensor through an edge computing node; performing space-time alignment on the processed sensor data and the equipment operation state parameters; establishing a digital twin model library on the cloud server; and performing mode comparison on the real-time sensor data stream and the fault features in the digital twin model library through a fuzzy matching algorithm, and when the vibration frequency domain features have amplitude sudden change exceeding a baseline value in a preset frequency band, the temperature change rate exceeding a preset threshold and the current harmonic distortion rate exceeding a set proportion, triggering a grading early warning signal. According to the invention, the accuracy, the real-time performance and the intelligent level of operation and maintenance of the building equipment can be obviously improved.

The invention discloses an engineering machinery fault prediction and intelligent maintenance method based on deep learning, and belongs to the technical field of intelligent operation and maintenance of engineering machinery. The method comprises the following steps: firstly, performing timestamp synchronization and feature enhancement on multi-source sensor data to generate a space-time alignment tensor; fusing the image and time sequence features through a multi-modal feature distillation network, and constructing a cross-modal unified feature vector; thirdly, calculating a fault probability and residual life distribution, and constructing a Markov decision model in combination with a resource state; and finally, dynamically optimizing the maintenance instruction by using deep reinforcement learning, and continuously updating the model through closed-loop feedback. According to the method, early-stage accurate prediction of the fault and dynamic optimization of the maintenance strategy are realized, the problems of inaccurate prediction, decision lag, resource waste and the like in a traditional method are effectively solved, and the availability rate and the maintenance economy of equipment are remarkably improved.

The invention provides a medical equipment fault detection method and system based on machine learning, and relates to the technical field of fault detection. The method comprises the steps of monitoring motor start-stop division operation cycles, synchronously collecting and processing vibration, frequency difference and voltage signals, calculating characteristic difference to generate a dynamic sequence, extracting a non-convergence trend and phase deviation, fusing voltage and frequency fluctuation characteristics, constructing a threshold rule to judge abnormity, and outputting a light fault early warning signal. According to the invention, through precise monitoring of motor start-stop nodes, period division, collection of vibration, frequency and voltage signals, combination of time alignment and noise filtering, improvement of signal quality, cross-period calculation of characteristic difference, and extraction of non-convergence trend and phase offset, the ability of capturing tiny anomalies is enhanced; according to the method, multi-dimensional indexes such as voltage fluctuation and frequency deviation are fused, a threshold judgment mechanism in a continuous window is constructed, stable identification and early warning of equipment light faults are achieved, and the detection accuracy and response timeliness are improved.

The invention relates to the technical field of sensors, and discloses a detection and management method and system for purification equipment. The method comprises the following steps: carrying out three-phase coupling dynamic modeling on a gas phase flow state, a particle phase movement track and a sensor phase interface characteristic in the purification equipment to obtain a gas mass transfer characteristic parameter; performing response characteristic analysis on a real-time detection signal of a sensor array in the purification equipment to obtain a sensor calibration compensation parameter; based on the sensor calibration compensation parameters and the gas mass transfer characteristic parameters, deposition kinetics analysis of pollutants on the surface of the sensor is carried out, and equipment operation state evaluation data are obtained; and performing spatial layout optimization on the sensor array according to the equipment operation state evaluation data and the gas flow consistency constraint condition to obtain an optimal layout scheme and generate a staged adaptive management control instruction. According to the invention, self-adaptive adjustment and intelligent switching of the sensor management strategy are realized, and the detection performance and reliability of the purification equipment sensor system are significantly improved.

The invention belongs to the technical field of fault early warning, and discloses a heater fault early warning method and system based on current analysis. Comprising the steps that transient current signals of all heating wire loops in the copper bush heater are collected in real time, harmonic characteristic parameters are extracted, and a dynamic current fingerprint database and a multi-dimensional harmonic characteristic spectrum are generated respectively; dynamically calculating the contact impedance between each heating wire and the copper sleeve groove according to the dynamic current fingerprint database and the multi-dimensional harmonic characteristic spectrum; the current distribution path of each heating wire in the copper sleeve groove is restored in real time, the abnormal contact form of each heating wire is identified, the falling risk probability and falling risk strength of each heating wire are dynamically predicted by fusing the falling precursor signals of each heating wire detected in real time, and a grading early warning strategy report is generated in real time; according to the invention, fault early warning can be converted into preventive maintenance from post-maintenance, the intelligent level of operation and maintenance of the heater is improved, and the continuity and safety of production are guaranteed.

The invention relates to the technical field of bridge safety monitoring, and comprises a bridge safety monitoring analysis system based on big data, and the system comprises a data collection processing module, a manifold feature dimension reduction module, an entropy change partition evaluation module, a health state dynamic analysis module, and a bridge safety early warning module. According to the method, a bridge measuring point topological relation matrix is established, measuring point spatial distribution characteristics are analyzed, spatial relevance of a bridge structure is reflected, measuring point local curvatures are calculated, a local curvature matrix is established, a neighborhood similarity matrix is combined, low-dimensional projection transformation is executed, local topological consistency of data is kept in the dimensionality reduction process, an entropy change partition matrix is established, and the spatial relevance of the bridge structure is reflected. The health states of different structural parts of the bridge are subjected to differential analysis based on mechanical characteristics, the entropy change trend and damage probability of a measuring point are calculated, and the overall safety state level of the bridge is judged through a damage risk threshold, so that the local damage severity of the bridge can be comprehensively considered in safety assessment, and the stability and adaptability of safety state judgment are improved.

The invention discloses an on-line monitoring and fault early warning system for the running state of a dynamic compression-shear testing machine, belongs to the technical field of fault diagnosis, and aims to solve the problems of poor adaptability to multiple motion modes, lagging fault early warning and fuzzy fault positioning in the prior art. According to the system, core collaborative link fault sensitive point monitoring parameters are matched according to a current motion mode, a dynamic threshold value is generated to construct a fault judgment threshold value system, real-time data subjected to cyclic division are collected and loaded to generate a time sequence data set, a reference candidate range state is judged based on the threshold value, and a fault judgment reference library is constructed; and executing deviation analysis prediction trend through the reference library, generating a compensation instruction, identifying potential faults in combination with a threshold value, an actual value and a prediction value, calculating a link fault probability, and completing diagnosis. According to the invention, monitoring adaptability and accuracy can be improved, compensation in advance and accurate fault early warning are realized, and stable operation and test accuracy of equipment are guaranteed.

The invention provides a fault prediction method and device for an energy storage liquid cooling system, relates to the technical field of artificial intelligence and data processing, and can better capture characteristics in different fault modes by dynamically analyzing local noise level and time scale change of monitoring data and performing adaptive segmentation processing on original data. And aiming at the characteristics of each segment of data, the denoising strategy is dynamically adjusted, so that noise can be effectively removed, and useful information can be reserved to the greatest extent. According to the method, the denoising process is accurately controlled, weak but key fault features can be accurately extracted even in a high-noise environment, and the method is crucial for improving the sensitivity and specificity of a fault diagnosis system.

The invention belongs to the technical field of photo-thermal performance detection of building envelope structures, and particularly relates to a light-transmitting roof photo-thermal performance detection device under various meteorological conditions and a detection method of the light-transmitting roof photo-thermal performance detection device. The invention discloses a light-transmitting roof photo-thermal performance detection device under various meteorological conditions. The device comprises a heat metering device, an air supply and return system, a water supply and return system, a control system and an external environment simulation device, wherein the external environment simulation device comprises an external environment box and an artificial light source; the light-transmitting roof test piece is installed on the lower side wall of the external environment box, and the external environment simulation device and the spraying device are used for simulating the environment conditions of different steady-state meteorological conditions and dynamic meteorological conditions under the spraying condition; the spraying device comprises a nozzle; a first light source radiation intensity sensor, a second light source radiation intensity sensor and a first illuminance sensor are further arranged in the outer environment box. According to the invention, the photo-thermal performance of the light-transmitting roof under the spraying condition can be accurately evaluated.

The application provides a large-span bridge substructure vehicle-fire wind force real-time hybrid test platform and implementation method, the platform comprises a physical loading and transmission system, a T-shaped wind-fire coupling simulation system, a modularized prying integrated system, an intelligent control and calculation system and a multi-dimensional high-precision monitoring system. The method realizes real-time pre-test of the boundary displacement / force of the test piece through the construction of a numerical substructure fast calculation model based on LSTM and converts the boundary displacement / force into a loading instruction to drive the through core jack to realize thermal force coupling interactive loading; the T-shaped wind-fire coupling simulation system reproduces the real vehicle fire scene under the influence of the environmental wind. The test platform supports independent loading of any component of the cable system, can execute open space self-defined fire test and closed space standard fire test, and accurately analyzes the cable failure mode, critical temperature threshold and damage evolution law under fire through the multi-dimensional high-precision monitoring system, thereby providing a high-fidelity test platform and analysis method for fire resistance design of large-span bridges.

The invention relates to the technical field of artificial intelligence and robots, and discloses a robot health state authentication method and system based on multi-dimensional fusion, and the method comprises the steps: collecting multi-source data, and carrying out the standardization preprocessing; 17 health dimensions are estimated based on an algorithm model; fusing the health dimensions to generate a comprehensive health index; authenticating the health state, and generating a health authentication report; reporting and coding, and storing to a block chain; the system comprises a multi-source data acquisition module, a dimension calculation engine, a correlation analysis module, a dynamic authentication generator, a block chain evidence storage module and an application interface layer. According to the method, a 17-dimensional health index system is constructed, hardware multiplexing is realized by applying an algorithm model based on current, speed, vision, network data and the like, an expensive physical sensor is effectively replaced, and more comprehensive health data can be obtained by matching with a chemical risk inversion model based on vision and network data to generate an environmental chemical index.

The invention discloses an electronic function test system of a high-voltage PTC (Positive Temperature Coefficient) electric heater, which relates to the technical field of PTC electric heaters and is characterized in that extreme temperature, humidity and voltage disturbance data are acquired in a programmable multi-dimensional environment simulation cabin, and the temperature deviation of the high-voltage PTC electric heater is predicted by using a multi-channel fusion and compensation model; according to the method, the high-voltage PTC electric heater is used as a power source, the output power is adjusted in a real-time feedforward and feedback mode in the self-adaptive controller, finally, through dynamic calibration and closed-loop verification, a model and gain parameters are optimized in multiple scenes, a complete system which can deal with large environmental changes and is high in temperature control precision is formed, the stability and safety of the high-voltage PTC electric heater under the severe working condition are remarkably improved, and the high-voltage PTC electric heater is suitable for being used in the field of electric heaters. And overshoot and lag are reduced, the energy utilization efficiency is improved, the system is suitable for industrial heating, vehicle warm air and other scenes needing high reliability and quick response, and the maintenance cost can be effectively reduced.

The invention provides a bridge structure damage detection method based on image processing and CNN-LSTM, relates to the field of graphic image processing, and solves the problems that bridge vibration monitoring in the prior art is time-consuming, labor-consuming, high in cost and limited in precision, and adopts the technical scheme that the method comprises the steps of obtaining a recorded vibration video of a to-be-detected bridge structure; extracting the vibration video frame by frame to obtain bridge image data; detecting angular points by using a Harris algorithm, screening image data, and calculating vertical displacement of each feature point one by one; using a Lucas-Kanade optical flow algorithm to calculate the motion vectors of the feature angular points of the adjacent frame images; calculating the actual vibration displacement of the bridge; and carrying out standardization processing on the actual vibration displacement to obtain a two-dimensional matrix, inputting the two-dimensional matrix into the trained CNN-LSTM model, and outputting a final identification result of each type of damage. According to the scheme of the invention, high-precision bridge vibration displacement can be calculated, and efficient, accurate and high-precision bridge damage detection can be realized through the CNN-LSTM damage identification model.

The invention belongs to the field of data processing, and provides an equipment comfort measurement fault identification method, system and equipment, and the method comprises the steps: obtaining a plurality of modal sequences which are consistent with a sampling signal segment in length and are aligned in bit sequence through variational modal decomposition; traversing each sampling signal segment according to a bit sequence to form a signal segment extended wave; extracting the peak value of the signal segment extended wave and the characteristic of the bit sequence of the peak value to generate an extended wave ratio sequence; the difference between the wave ratio sequence of each sampling signal segment and the original sequence is used as the wave ratio fall; and screening each sampling signal segment according to the wave ratio fall by using the median of the wave ratio fall of all the sampling signal segments, and identifying a signal comfort abnormal segment. The method is not sensitive to amplitude fluctuation of random noise, but is more sensitive to envelope impact and bit sequence displacement of a specific frequency band, can prompt slight problems such as guide shoe clearance, guide rail joint impact and abnormal transmission of a traction machine in the early stage, and is beneficial to preventive maintenance.

The invention relates to the technical field of water electrolysis hydrogen production, and discloses a method and system for evaluating the operation performance of an alkaline water electrolysis hydrogen production device, and the method comprises the steps: 1, deploying a distributed sensor network, collecting the operation data of an electrolytic cell, building a multi-physical field coupling model, and calibrating a model parameter matrix; and step 2, constructing an electrode damage evolution model based on the model parameter matrix, processing acoustic emission signals acquired by the sensor network, extracting crack characteristic parameters, calculating real-time damage degree, and generating graded early warning signals. According to the technical scheme, the distributed sensor network is adopted to collect the multi-physical field data, and the coupling model is established to calibrate the parameter matrix, so that the technical effect of a multi-dimensional parameter dynamic coupling relation is achieved; the defects of large energy efficiency evaluation deviation and low reliability caused by neglecting of multi-field interaction in a traditional method are overcome.

The invention relates to the technical field of fault diagnosis, in particular to a hydraulic engineering equipment fault diagnosis method and system.The hydraulic engineering equipment fault diagnosis method comprises the following steps of collecting water inlet and outlet flow velocity data of a water pump to judge direction reversal, extracting rotating speed and pressure trend to collect linkage characteristics, analyzing starting and stopping behaviors to calibrate abnormal sections, and integrating multi-parameter generation structural block screening combination characteristics; arranging the structure blocks to form a continuous path, and classifying the matching mode to generate a fault trend diagnosis path set. According to the method, the direction change of the adjacent sampling points in the flow velocity data of the water inlet and outlet channel of the water pump is continuously judged, the section with transition instability in the local flow velocity structure is recognized, fault symptoms of multiple dimensions are merged into recognizable structural blocks, the overall integrity of feature recognition is improved, and the accuracy of feature recognition is improved. The fault diagnosis process is improved from single-parameter abnormal identification to composite matching of a multi-dimensional behavior mode, and the method has the capability of tracking and positioning the whole process of the operation trend of hydraulic engineering equipment.

The present invention relates to a collaborative robot system used in an industrial environment where workers collaborate. The collaborative robot system comprises: a collaborative robot body having a multi-joint structure; a control unit for controlling each joint or auxiliary shaft; one or more sensors for detecting the operational state of the robot in real time; and a status diagnosis and response module for diagnosing the state of the robot on the basis of detected data and controlling the operation of the robot accordingly. Specifically, the system is configured to enable the robot to autonomously perform actions such as deceleration, stopping, and recovery in response to various state changes occurring during work, and to perform self-learning and trajectory optimization on the basis of data accumulated through repetitive tasks. Furthermore, the system includes a graphical user interface (GUI) for intuitive user interaction, which is linked to a digital twin-based virtual simulation environment, enabling presetting and modification of work paths. When multiple collaborative robots are operated together, the system enables task synchronization, path collision avoidance, and sharing of status information among the robots, and may be connected to an external control server or a cloud-based control system to allow integrated management of the entire workflow.

To provide a method for predicting a strain distribution map and determining damage based on deep learning.SOLUTION: The present invention comprises: a step S1 of establishing an image dataset of finite element analysis results for strain distribution map prediction; a step S2 of building a deep learning model for strain distribution map prediction based on a DeepLabv3+ network and performing learning and validation; a step S3 of establishing a dataset for damage determination of a structural analysis model and performing preliminary processing and data enhancement operations; a step S4 of building a binary classification deep learning model for damage determination based on a convolutional neural network and performing learning and validation for transition learning; and a step S5 of performing an interpretability analysis on the learned binary classification structural analysis model and outputting a region in an image that more contributes to classification.SELECTED DRAWING: Figure 1

The invention discloses a load simulation loading system based on an engineering mechanical power system and a control method thereof, and relates to the technical field of mechanical power system simulation. The load simulation loading system comprises an engineering machinery power system, a load loading system, a whole vehicle control unit and an upper computer platform. The engineering machinery power system comprises a first oil tank, a main pump, a pilot pump, a pilot control handle, a signal control valve and a multi-way valve. The load loading system comprises a three-position four-way reversing valve, a proportional overflow valve, a first proportional throttle valve, a second proportional throttle valve, an electromagnetic valve, a third proportional throttle valve and a variable pump, wherein the three-position four-way reversing valve, the proportional overflow valve and the first proportional throttle valve are connected to the multi-way valve, the second proportional throttle valve and the electromagnetic valve are connected to an outlet of the proportional overflow valve, and the third proportional throttle valve is connected between the second proportional throttle valve and the second oil tank. And the three-position four-way reversing valve is constructed to enable an inlet of the proportional overflow valve to be communicated with one of the oil port A and the oil port B of the multi-way valve in a switching manner, and an outlet of the proportional overflow valve to be communicated with the other one of the oil port A and the oil port B in a switching manner.

The invention belongs to the technical field of photo-thermal performance detection of building envelope structures, and particularly relates to a device and a method for detecting a solar heat gain coefficient under dynamic meteorological conditions of door and window curtain walls. The invention discloses a device for detecting a solar heat gain coefficient under a door and window curtain wall dynamic meteorological condition, and the device comprises an outdoor environment simulation hot chamber which comprises a solar radiation simulation system, a wind speed simulation system, and a temperature and humidity simulation system; an environmental space; the heat metering box and the outdoor environment simulation hot chamber are arranged in the horizontal direction, and the heat metering box is located in the environment space; a tested piece is arranged at the joint interface of the outdoor environment simulation hot chamber and the heat metering box along the vertical direction; the solar radiation simulation system adjusts the solar radiation intensity according to the received dynamic meteorological conditions, the wind speed simulation system adjusts the wind speed according to the received dynamic meteorological conditions, and the temperature and humidity simulation system adjusts the temperature and humidity according to the received dynamic meteorological conditions. According to the invention, the detection precision of the adjustable light-transmitting enclosure structure can be improved.

The invention discloses a heat pipe heat exchanger fault real-time monitoring system and method based on multi-source data fusion, and the system comprises a multi-source data collection module which is used for collecting temperature data, pressure data, flow data, vibration data and image data in the operation process of a heat pipe heat exchanger; the invention relates to the technical field of heat pipe heat exchanger fault monitoring. According to the heat pipe heat exchanger fault real-time monitoring system and method based on multi-source data fusion, multi-source data such as temperature, pressure, flow, vibration and images are creatively fused, and compared with traditional single sensor monitoring, operation state information of the heat pipe heat exchanger can be captured in an all-around mode. Various sensors are arranged at key positions, such as temperature sensors at the inlet, the outlet, the pipe wall and the like, so that the internal temperature distribution of the heat exchanger can be comprehensively mastered; the pressure sensor and the flow sensor can reflect the flowing state of fluid; visual basis is provided for diagnosis of mechanical faults and external states by vibration and image data.

According to the equipment state recognition method and device, the equipment and the storage medium provided by the embodiment of the invention, the multi-modal data in the operation process of the target equipment is acquired, and the multi-modal data comprises the image data, the text data and the temperature data; performing feature extraction and embedded coding on the multi-modal data through a feature extraction model to obtain embedded vectors of all modals, and aligning the embedded vectors of all modals to a unified target semantic embedding space; through a modal attention mechanism, the fusion weight of each modal is adjusted according to the state feature of the target device, and the embedding vectors of each modal are fused according to the fusion weight of each modal to construct a multi-modal fusion vector; and recognizing the state of the target equipment and / or performing abnormal risk early warning through a recognition model according to the multi-modal fusion vector. By introducing a modal attention mechanism, dynamic weighted fusion of image, text and temperature modals is realized, the accuracy and generalization ability of state recognition are remarkably improved, and the accuracy and robustness of the model under complex working conditions are improved.

The invention relates to the field of thermal fault early warning of machine room inspection robots, in particular to a multi-modal sensing thermal fault early warning method of a machine room inspection robot, which comprises the following steps: acquiring infrared thermal image, acoustics, vibration and gas concentration data through a multi-modal sensor array, and generating voxelization temperature distribution; according to the method, temperature features are extracted through wavelet transform and a fractional Brownian motion algorithm and are fused into an equipment global temperature feature set; constructing noise, vibration and gas concentration tower extraction auxiliary features, and combining non-extensive entropy and a graph attention model to be fused into a high-dimensional fault feature pool; building a fiber bundle model by taking the machine room topology as a base space, and locating a thermal fault after optimizing parameters by a simulated annealing algorithm; and finally, a dynamic thermodynamic diagram and the like are generated through digital twinning, so that accurate early warning is realized.

The invention discloses an intelligent diagnosis and prediction method and system for mine equipment faults based on deep learning, which is applied to the technical field of mine equipment fault diagnosis, and comprises the following steps: obtaining operation data of mine equipment, and carrying out data preprocessing; performing feature extraction on time domain, frequency domain and time-frequency combination on the preprocessed data, taking each time step length as a feature vector, inputting the normalized features of the data into an improved LSTM neural network for training, and obtaining an improved LSTM-based mine equipment fault intelligent diagnosis and prediction model; and inputting to-be-detected data to the mine equipment fault intelligent diagnosis and prediction model to obtain a mine equipment fault intelligent diagnosis and prediction result. According to the invention, the key features in the operation data of the mine equipment are effectively and automatically extracted, the influence of artificial subjective factors is avoided, the accuracy and timeliness of fault diagnosis are effectively improved, and safe, stable and efficient operation of coal mine production is guaranteed.

The invention provides a water chilling unit performance prediction method based on an improved physical information neural network. The method comprises the following steps: step 10, constructing a neural network model; step 20, selecting to obtain a neuron variable, and embedding the neuron variable into the neural network model as a physical neuron to obtain a physical neuron embedded physical information neural network model; step 30, introducing a loss function into the physical neuron embedded physical information neural network model; 40, carrying out data balance processing on the original data set, and screening to obtain a training set; step 50, utilizing the training set to train the physical neuron embedded physical information neural network model to obtain a water chilling unit performance prediction model; and step 60, predicting the performance of the water chilling unit by using the water chilling unit performance prediction model. According to the improved water chilling unit performance prediction method based on the physical information neural network, high-precision prediction of the performance of the water chilling unit is achieved, and the generalization ability and the model interpretability are improved.