Motor temperature monitoring and regulation method and system based on intelligent control and storage medium

Abstract

The invention relates to the technical field of motor temperature intelligent control, and particularly discloses a motor temperature monitoring and regulation method and system based on intelligent control and a storage medium, and the method comprises the steps: carrying out data collection circulation: judging whether a sampling period is reached, if yes, reading multi-mode sensing data, and outputting a calibrated temperature value T, and if not, outputting a calibrated temperature value T; if yes, vibration noise is eliminated, the environment temperature is compensated, multimode sensing data is added, and then the calibrated temperature value T is output; intelligent analysis is conducted on the calibrated temperature value T obtained through data collection circulation, if the temperature is normal, data analysis and prediction are conducted, and if the temperature is abnormal, the calibrated temperature value T is analyzed, fault self-diagnosis is conducted, abnormal reasons are judged, the system is controlled to conduct self-adjustment, and temperature early warning is conducted. The problems that in the prior art, monitoring is delayed, regulation and control are rigid, and maintenance is passive, so that energy efficiency is wasted, the failure rate is high, and operation and maintenance cost rises are solved; and a real-time temperature monitoring means and an intelligent control system are lacked.

Description

Technical Field

[0001] The present application relates to the technical field of motor temperature intelligent control, and specifically to a motor temperature monitoring and regulation method, system and storage medium based on intelligent control. Background Art

[0002] Motors are the core power source for mining machinery, power equipment, metallurgical production lines, and other fields. Their operating temperature directly impacts their efficiency, lifespan, and safety. Prolonged high temperatures can lead to insulation aging, lubrication failure, winding burnout, and other failures, resulting in unplanned downtime and even accidents. According to statistics, approximately 35% of motor failures are directly related to overheating, with bearing overheating accounting for 22% of all failures.

[0003] Analysis of pain points of existing technologies:

[0004] 1. Lack of real-time temperature monitoring means:

[0005] Traditional motors mostly rely on regular manual inspections or simple mechanical temperature control switches, which are unable to capture dynamic changes in motor temperature in real time and make it difficult to detect sudden temperature rises (such as overheating caused by insufficient lubrication or sudden load changes) in a timely manner.

[0006] 2. Passive maintenance and high failure rate:

[0007] Existing equipment is typically repaired after a failure occurs, resulting in frequent unplanned downtime and high repair costs.

[0008] Disadvantages of existing technology:

[0009] 1. Temperature monitoring has serious lag:

[0010] Relying on manual inspections or mechanical temperature control switches has high detection delays and cannot capture sudden temperature rises in real time (such as instantaneous overheating caused by bearing jamming and lubrication failure).

[0011] 2. Maintenance costs surge:

[0012] Sudden bearing failures and other problems can lead to cascading equipment shutdowns, and the cost of a single repair can reach 15%-30% of the equipment value.

[0013] 3. Lack of predictive maintenance capabilities:

[0014] Existing solutions use passive fault diagnosis, relying on post-event temperature alarms (such as shutdown when exceeding 105°C), and lack early warning of temperature trends.

[0015] 4. Lack of data management:

[0016] Data silos exist. Temperature data isn't integrated with multi-dimensional signals like vibration and current, making it impossible to build a health assessment model. The lack of historical temperature data accumulation and the ability to analyze temperature trends and predict lifespans means equipment maintenance relies on post-event repairs, increasing the frequency of unplanned downtime and preventing predictive maintenance through trend analysis. Summary of the Invention

[0017] The purpose of this application is to provide a motor temperature monitoring and control method, system and storage medium based on intelligent control, so as to solve the three major shortcomings of the existing technology: monitoring delay, rigid control, and passive maintenance, which lead to energy waste, high failure rate, and rising operation and maintenance costs; and the lack of real-time temperature monitoring means and intelligent control systems.

[0018] To achieve the above objectives, the present invention provides a method for monitoring and controlling motor temperature based on intelligent control, comprising:

[0019] Performing a data acquisition cycle specifically includes: determining whether the sampling period has been reached; if so, reading the multi-mode sensor data and outputting the calibrated temperature value T; if not, eliminating vibration noise, compensating for the ambient temperature, and appending the multi-mode sensor data before outputting the calibrated temperature value T;

[0020] The calibrated temperature value T obtained in the data acquisition cycle is intelligently analyzed: if the temperature is normal, data analysis and prediction are performed; if the temperature is abnormal, the calibrated temperature value T is analyzed, fault self-diagnosis is performed, the cause of the abnormality is determined, the control system performs self-adjustment, and a temperature warning is issued.

[0021] Optionally, the data analysis and prediction specifically includes:

[0022] The calibrated temperature value T is uploaded to the data cloud platform, a temperature-time relationship chart is drawn, and the constructed prediction model TTNet is called. The time-temperature relationship sequence in the relationship chart is input. After analysis, the future time-temperature relationship index CHI is output. If the CHI value is abnormal, a preventive maintenance work order is generated and an alarm is sent. If the CHI value is normal, the data collection loop returns.

[0023] Optionally, after analyzing the calibrated temperature value T, the method further includes:

[0024] Calculate the temperature rise rate. If the temperature rise is too fast, the control system will self-adjust, generate a maintenance report, and send an alarm. If the temperature rise rate reaches the shutdown threshold, an emergency shutdown will be initiated, power will be cut off, an alarm will be sounded, pre-fault data will be saved, and a fault analysis report will be generated.

[0025] Optionally, performing temperature warning specifically includes:

[0026] By analyzing environmental parameters such as ambient temperature and relative humidity, a dynamic mapping relationship between environmental parameters and warning thresholds is established, and the warning temperature is adjusted in real time. When the temperature reaches the general temperature warning, the control system self-adjusts, generates a maintenance report, and sends an alarm. When the temperature reaches the emergency shutdown threshold, the power is cut off, an alarm is triggered, pre-fault data is saved, and a fault analysis report is generated.

[0027] Optionally, the constructed prediction model TTNet is called, the time and temperature relationship sequence of the relationship chart is input, and after analysis, the future time-temperature relationship index CHI is output, specifically including:

[0028] The input data of the prediction model TTNet includes:

[0029] Time series features: including time and temperature relationship sequence,

[0030] Auxiliary features: including vibration signal spectrum, motor current load, ambient temperature and humidity;

[0031] The output of the prediction model TTNet includes:

[0032] The predicted temperature and temperature rise rate for the next m time steps.

[0033] Optionally, the network architecture of the constructed prediction model TTNet includes:

[0034] The spatiotemporal feature encoding layer is used to fuse multimodal input data and extract spatiotemporal correlation features;

[0035] Multi-scale temporal attention module for dynamically weighting features at different time scales;

[0036] Thermodynamic residual constraint layer, used to constrain network output through physical equations;

[0037] Online adaptive module, used to dynamically update model parameters based on real-time data;

[0038] The fusing of multimodal input data and extracting spatiotemporal correlation features specifically includes:

[0039] Temporal Convolutional Network: Uses dilated causal convolution to capture multi-scale temporal dependencies and output feature maps.

[0040] Thermodynamic physics encoder: Input: time-temperature relationship sequence and ambient temperature, calculate heat conduction residual, and embed the residual as a constraint term into the feature vector;

[0041] The dynamic weighting of different time-scale features specifically includes: calculating attention weights for the multi-scale features output by the temporal convolutional network, and outputting fused features;

[0042] The output of the network is constrained by physical equations, specifically including:

[0043] Define the loss function to include data error and physical residual, and control the strength of physical constraints through hyperparameters;

[0044] The dynamic updating of model parameters according to real-time data specifically includes:

[0045] When the prediction error of multiple consecutive time steps exceeds the threshold, incremental learning is triggered to use elastic weight solidification to prevent catastrophic forgetting.

[0046] Optionally, the triggering of incremental learning specifically includes:

[0047] Error monitoring: Real-time calculation of the mean absolute error between the predicted temperature and the measured temperature. If the mean absolute error for N consecutive time steps exceeds the threshold, incremental learning is triggered.

[0048] Data buffer pool: stores the temperature, vibration, and current time series data of the last K hours. It uses a ring buffer strategy, with new data overwriting old data, and the memory usage remains constant.

[0049] The elastic weight solidification method specifically includes:

[0050] Introduce a regularization term into the loss function:

[0051] ,

[0052] in:

[0053] : prediction loss of new data (MSE);

[0054] : Model parameters before incremental learning;

[0055] : Fisher information matrix diagonal elements, quantization parameters the importance of

[0056] : Regularization strength.

[0057] Optionally, the performing fault self-diagnosis specifically includes:

[0058] Fault knowledge base construction: By encoding typical temperature-related faults of motors, recording fault types and typical temperature rise characteristics; labeling historical fault cases to build a training set;

[0059] Fault self-diagnosis logic:

[0060] Online Fault Detection: Anomaly Scores calculate:

[0061] ,

[0062] in, : Historical baseline value of temperature standard deviation under current working conditions,

[0063] 、 : weight coefficient,

[0064] : Actual temperature value,

[0065] : Temperature prediction value,

[0066] : Normalized value of temperature rise rate,

[0067] Trigger condition: If If it is greater than the set value, it is determined to be a potential fault and the root cause analysis is started.

[0068] Fault pattern matching: extract key feature vectors from the output, calculate the similarity between the feature vectors and the historical fault features in the knowledge base, and return the corresponding fault code if the similarity is greater than the set value; otherwise, mark it as "unknown fault".

[0069] Unknown fault learning: Online clustering and cluster analysis are performed on "unknown fault" samples. If the number of samples in the new category is greater than 50, a temporary fault code is created and the data related to the temporary fault code is pushed to the operation and maintenance personnel, who then request labeling and add it to the knowledge base.

[0070] To achieve the above objectives, the present application also provides a motor temperature monitoring and control system based on intelligent control, including: a regulating mechanism, a multimodal sensing device, an intelligent control system, and a data cloud platform, wherein:

[0071] The multi-modal sensing device uploads the collected multi-modal sensing data such as motor temperature, environmental parameters, vibration spectrum, etc. to the intelligent control system;

[0072] The intelligent control system reads the sensor data and performs intelligent analysis, and then uploads compressed temperature history data, control logs, fault events and other data to the data cloud platform;

[0073] The data cloud platform is deployed with a prediction model TTNet for predicting future temperature trends and calculating the future time-temperature relationship index CHI;

[0074] The data cloud platform transmits cycle optimization suggestions, load adjustment strategies, and model parameter updates to the intelligent control system;

[0075] The intelligent control system updates various coefficients and adjusts the parameters of the motor body and the regulating mechanism;

[0076] The intelligent control system transmits control instructions to the motor body and the regulating mechanism, adjusts the motor state in real time, and maintains motor safety and temperature stability.

[0077] To achieve the above objectives, the present application also provides a computer storage medium having a computer program stored thereon, wherein the computer program implements the steps of the above method when executed by a machine.

[0078] The embodiments of the present application have the following advantages:

[0079] This method, combined with sensor technology and the Internet of Things (IoT), enables real-time monitoring, dynamic control, and fault warning of motor temperature. It is suitable for intelligent management of high-load motors in scenarios such as mining machinery, power equipment, and metallurgical production lines. It addresses the three shortcomings of existing technologies: delayed monitoring, rigid control, and passive maintenance, which lead to energy waste, high failure rates, and escalating operation and maintenance costs; and the lack of real-time temperature monitoring and intelligent control systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] To more clearly illustrate the embodiments of this application or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are merely illustrative, and those skilled in the art can, without inventive effort, derive other implementation drawings based on the provided drawings.

[0081] Figure 1 A module block diagram of a motor temperature monitoring and control system based on intelligent control provided in at least one embodiment of the present application;

[0082] Figure 2 A flow chart of a method for monitoring and regulating motor temperature based on intelligent control provided in at least one embodiment of the present application;

[0083] Figure 3 A flowchart of a method for constructing a prediction model TTNet for a motor temperature monitoring and control method based on intelligent control provided in at least one embodiment of the present application. DETAILED DESCRIPTION

[0084] The following specific embodiments illustrate the implementation of this application. Those familiar with the art can easily understand the other advantages and functions of this application from the content disclosed in this specification. Obviously, the described embodiments are part of the embodiments of this application, but not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0085] It should be noted that, in the claims and description of this application, the steps may be executed substantially in parallel or in reverse order under appropriate circumstances, depending on the functions involved.

[0086] In addition, the technical features involved in the different embodiments of the present application described below can be combined with each other as long as they do not conflict with each other.

[0087] Figure 1 This is a block diagram of a motor temperature monitoring and control system based on intelligent control, provided in at least one embodiment of the present application. The system includes:

[0088] Motor body, adjustment mechanism, multimodal sensing device, intelligent control system and data cloud platform.

[0089] 1. Motor body: core equipment, generates heat and mechanical output.

[0090] 2. Adjustment mechanism: cooling fan, lubrication mechanism, etc., to adjust the motor status.

[0091] 3. Multimodal sensing device: Multimodal sensing, real-time collection of multi-dimensional temperature data.

[0092] 4. Intelligent control system: Processes data, executes control strategies, and regulates heat dissipation and lubrication. It also provides real-time edge-side regulation and emergency response.

[0093] 5. Data cloud platform: performs long-term data analysis and predictive maintenance; stores historical data and manages clusters; and calculates the future time-temperature relationship index (CHI) using the predictive model TTNet (ThermoTemporal Net).

[0094] The four components constitute a fully closed-loop intelligent temperature control system consisting of perception, decision-making, execution, and optimization. The interaction between data flow and control flow is as follows:

[0095] 1. Uplink data flow (perception → decision → optimization):

[0096] The multi-modal sensing device uploads the collected multi-modal sensing data such as motor temperature, environmental parameters, vibration spectrum, etc. to the intelligent control system;

[0097] The intelligent control system reads the sensor data and performs intelligent analysis, and then uploads compressed temperature history data, control logs, fault events and other data to the data cloud platform;

[0098] The prediction model TTNet is deployed on the data cloud platform to predict future temperature trends and calculate the future time-temperature relationship index CHI.

[0099] 2. Downstream data flow (optimization → execution → feedback):

[0100] The data cloud platform transmits cycle optimization suggestions, load adjustment strategies, model parameter updates, etc. to the intelligent control system;

[0101] The intelligent control system updates various coefficients and adjusts the parameters of the motor body and regulating mechanism;

[0102] The intelligent control system transmits control instructions to the motor body and adjustment mechanism, adjusts the motor status in real time, and maintains motor safety and temperature stability.

[0103] Specifically, through a collaborative working mechanism, a control closed loop of multimodal sensing device → intelligent control system → motor body actuator and adjustment mechanism is realized to cope with various emergency scenarios; a long-term optimization closed loop of data cloud platform → intelligent control system → motor body is realized to perform predictive maintenance; two-way data synchronization is performed, and the intelligent control system locally stores data for a certain period of time, maintains basic control functions to cope with network outages, and automatically retransmits data after the network is restored to ensure analysis continuity.

[0104] An embodiment of the present application also provides a motor temperature monitoring and control method based on intelligent control, referring to Figure 2 , Figure 2 This is a flowchart of a method for monitoring and regulating motor temperature based on intelligent control provided in at least one embodiment of the present application. It should be understood that the method may also include additional boxes not shown and / or the boxes shown may be omitted, and the scope of the present application is not limited in this respect.

[0105] 1. System startup: system power-on initialization;

[0106] 2. System preparation: sensor calibration, loading temperature control parameters; establishing data cloud platform connection, access to intelligent control system;

[0107] 3. Perform a data acquisition cycle, specifically including: determining whether the sampling period (e.g., 0.1s) has been reached; if so, reading the multi-mode sensor data and outputting the calibrated temperature value T; if not, eliminating vibration noise, compensating for the ambient temperature, appending the multi-mode sensor data, and then outputting the calibrated temperature value T;

[0108] 4. Intelligent control: Intelligent analysis of the calibrated temperature value T obtained in the data acquisition cycle:

[0109] ① If the temperature is normal, data analysis and prediction are performed. In some embodiments, this specifically includes: uploading the calibrated temperature value T to the data cloud platform, having the system draw a temperature-time relationship chart, calling the constructed prediction model TTNet (ThermoTemporal Net), inputting the time-temperature relationship sequence in the relationship chart, and outputting the future time-temperature relationship index CHI (ChronoHeat Index) after analysis. If the CHI value is abnormal, a preventive maintenance work order is generated and an alarm is issued. If the CHI value is normal, the data collection cycle is returned;

[0110] ② If the temperature is abnormal, the calibrated temperature value T is analyzed, and fault self-diagnosis is performed to determine the cause of the abnormality, such as motor overload, lack of lubricating oil, end of motor life or other emergencies. The control system performs self-adjustment, generates a maintenance report, sends an alarm, and issues a temperature warning.

[0111] In some embodiments, after analyzing the calibrated temperature value T, the method further includes: calculating the temperature rise rate. If the temperature rise is too fast, the control system performs self-adjustment, generates a maintenance report, and sends an alarm. If the temperature rise rate reaches the shutdown threshold, an emergency shutdown is performed, the power is cut off, an alarm is issued, pre-fault data is saved, and a fault analysis report is generated.

[0112] In some embodiments, the temperature warning specifically includes: establishing a dynamic mapping relationship between environmental parameters and warning thresholds through analysis of environmental parameters such as ambient temperature and relative humidity, and adjusting the warning temperature in real time; when the temperature reaches the general temperature warning, the control system performs self-adjustment, such as increasing the power of the cooling system, increasing the fan speed, starting the backup lubrication system, limiting the load on the motor, etc., and generating a maintenance report and sending an alarm; when the temperature reaches the emergency shutdown threshold, the power is cut off, an alarm is issued, the pre-fault data is saved, and a fault analysis report is generated.

[0113] 5. Loop monitoring: If the CHI value is normal, return to the data acquisition loop.

[0114] 6. System termination: The system receives a shutdown command, shuts down all actuators, and ends.

[0115] refer to Figure 3 In some embodiments, the calling of the constructed prediction model TTNet inputs the time and temperature relationship sequence of the relationship chart, analyzes it and outputs the future time-temperature relationship index CHI, specifically including:

[0116] 1. Model input and output

[0117] ① Input data:

[0118] Time series characteristics: time and temperature relationship sequence T t-n , T t-n+1 , T t-n+2 ,...,T t , where the time window n is adjustable and can be set to a default value, such as n=72, and t represents time.

[0119] Auxiliary features: vibration signal spectrum, motor current load, ambient temperature and humidity.

[0120] ② Output result: Temperature prediction value T for the next m time steps ^ t+1 , T ^ t+2 ,...,T ^ t+m ;

[0121] Predicted temperature rise rate: .

[0122] 2. Network architecture design

[0123] (1) Spatio-Temporal Encoder

[0124] Function: Fuse multimodal input data and extract spatiotemporal correlation features.

[0125] Implementation method: ① Temporal Convolutional Network (TCN): Use Dilated Causal Convolution with dilation factor d=2 k (k=0, 1, 2), capture multi-scale temporal dependencies and output feature maps;

[0126] ②Thermodynamics and physics encoder:

[0127] Input: time and temperature relationship sequence T and ambient temperature T env ;

[0128] Calculate the heat conduction residual:

[0129]

[0130] : Temperature time derivative,

[0131] : thermal diffusivity,

[0132] : Temperature space Laplacian operator,

[0133] The residual is embedded into the feature vector as a constraint term.

[0134] (2) Multi-Scale Temporal Attention

[0135] Function: Dynamically weight different time scale features to enhance sensitivity to sudden temperature rise.

[0136] Implementation method: Calculate attention weights for the multi-scale features (original sampling, short-time aggregation, long-time aggregation) output by the temporal convolutional network:

[0137]

[0138] : query matrix, feature representation of the current target,

[0139] : key matrix, key values ​​of historical features (matching query requirements),

[0140] : value matrix, which actually carries the feature information (output after weighted aggregation),

[0141] : Scaling factor to prevent the gradient from disappearing due to excessive dot product,

[0142] : Attention weight, the importance of features at each time step,

[0143] Output fused features.

[0144] (3) Thermodynamic Residual Layer

[0145] Function: Constrain network output through physical equations to improve model extrapolation capabilities.

[0146] Implementation method: Define the loss function to include data error and physical residual:

[0147]

[0148] : predicted temperature value,

[0149] : Measured temperature value,

[0150] : prediction step length,

[0151] : number of physical residual samples,

[0152] : physical residual,

[0153] : regularization coefficient,

[0154] The hyperparameter λ controls the strength of the physical constraint.

[0155] (4) Online Adaptation Module

[0156] Function: Dynamically update model parameters based on real-time data to adapt to scenarios such as equipment aging and environmental changes.

[0157] Implementation method: When the prediction error of multiple consecutive time steps exceeds a threshold (such as MAE>1.5℃), incremental learning is triggered:

[0158]

[0159] : old model parameters,

[0160] : learning rate,

[0161] : Gradient operator

[0162] : current loss,

[0163] Using Elastic Weight Curing (EWC):

[0164]

[0165] : model parameters,

[0166] : old model parameters,

[0167] : regularization coefficient,

[0168] Preventing catastrophic forgetting.

[0169] In some embodiments, triggering incremental learning specifically includes:

[0170] 1. Error monitoring: Real-time calculation of the MAE (mean absolute error) between the predicted temperature and the measured temperature:

[0171]

[0172] If the mean absolute error of N consecutive time steps (such as the default N=5) exceeds a threshold (such as 1.5°C), incremental learning is triggered.

[0173] 2. Data buffer pool: Stores time series data (temperature, vibration, current) for the last K hours (e.g., the default K=24). A ring buffer strategy is used, with new data overwriting old data, and memory usage remains constant.

[0174] In some embodiments, the elastic weight solidification specifically includes:

[0175] In order to protect existing knowledge from being covered, a regularization term is introduced into the loss function:

[0176]

[0177] in:

[0178] : prediction loss of new data (MSE);

[0179] : Model parameters before incremental learning;

[0180] : Fisher information matrix diagonal elements, quantization parameters the importance of

[0181] : Regularization strength.

[0182] In some embodiments, the performing fault self-diagnosis specifically includes:

[0183] By learning the mapping relationship between temperature anomalies and fault types through historical data, we can achieve an upgrade from "predicting temperature rise" to "locating the root cause".

[0184] 1. Fault knowledge base construction: By encoding typical temperature-related faults of the motor, recording the fault type, i.e., typical temperature rise characteristics; labeling historical fault cases (single fault and compound fault), and building a training set ;

[0185] 2. Fault self-diagnosis logic:

[0186] ① Online fault detection: anomaly score calculate:

[0187]

[0188] : Historical baseline value of temperature standard deviation under current working conditions,

[0189] 、 : weight coefficient,

[0190] : Actual temperature value,

[0191] : Temperature prediction value,

[0192] : Normalized value of temperature rise rate,

[0193] Trigger condition: If If it is greater than the set value, it is determined to be a potential fault and the root cause analysis is initiated.

[0194] ② Fault pattern matching: Extract key feature vectors from the output and calculate the similarity between the feature vectors and historical fault features in the knowledge base. If the similarity is greater than the set value, the corresponding fault code is returned; otherwise, it is marked as "unknown fault";

[0195] ③Unknown fault learning: Perform online clustering (Mini-Batch K-Means) on "unknown fault" samples and perform cluster analysis. If the number of samples in the new category is greater than 50, create a temporary fault code (such as UF01). Push the data related to the temporary fault code to the operation and maintenance personnel, request labeling, and then add it to the knowledge base.

[0196] In the embodiment of the present application, the future time-temperature relationship index CHI (ChronoHeat Index):

[0197]

[0198] : Normalized value of the predicted temperature series output by TTNet;

[0199] τ: time decay factor (related to the heat capacity of the device);

[0200] : Normalized value of temperature rise rate.

[0201] Physical meaning:

[0202] The higher the CHI value, the greater the temperature rise and the more dramatic the change in the future, and the higher the risk level of the equipment;

[0203] The logarithmic function compresses the impact of extreme temperature rise rates to avoid over-sensitivity of the indicator.

[0204] The present application may be a method, apparatus, system and / or computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present application.

[0205] A computer-readable storage medium can be a tangible device that can hold and store instructions used by an instruction execution device. Computer-readable storage media can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure within a groove on which instructions are stored, and any suitable combination thereof. As used herein, a computer-readable storage medium is not to be construed as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse passing through a fiber optic cable), or an electrical signal transmitted through an electrical wire.

[0206] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions to be stored in the computer-readable storage medium in each computing / processing device.

[0207] The computer program instructions used to perform the operations of the present application can be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The computer-readable program instructions can be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, the state information of the computer-readable program instructions is used to personalize an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), so that the electronic circuit can execute the computer-readable program instructions, thereby implementing various aspects of the present application.

[0208] Various aspects of the present application are described herein with reference to flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present application. It should be understood that each block of the flowcharts and / or block diagrams, and combinations of blocks in the flowcharts and / or block diagrams, can be implemented by computer-readable program instructions.

[0209] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, such that when these instructions are executed by the processing unit of the computer or other programmable data processing device, a device is generated that implements the functions / actions specified in one or more blocks in the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium, where these instructions cause the computer, programmable data processing device, and / or other device to operate in a specific manner. Thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing various aspects of the functions / actions specified in one or more blocks in the flowchart and / or block diagram.

[0210] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device, so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0211] The flow charts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the system, method and computer program product according to multiple embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and the part of the module, program segment or instruction includes one or more executable instructions for realizing the logical function of the specification. In some alternative implementations, the functions marked in the box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented by a special hardware-based system that performs the function or action of the specification, or can be implemented by a combination of special hardware and computer instructions.

[0212] Note that, unless otherwise explicitly stated, all features disclosed in this specification (including any accompanying claims, abstracts, and drawings) may be replaced by alternative features that achieve the same, equivalent, or similar purposes. Therefore, unless explicitly stated otherwise, each feature disclosed is merely an example of a group of equivalent or similar features. Where used, further, preferably, further, and more preferably are simply the beginning of another embodiment based on the previous embodiment, and the content following further, preferably, further, or more preferably is combined with the previous embodiment as a complete construction of another embodiment. Several further, preferably, further, or more preferably settings following the same embodiment can be arbitrarily combined to form another embodiment.

[0213] Although the present application has been described in detail above using general descriptions and specific embodiments, it will be apparent to those skilled in the art that modifications or improvements may be made to the present application. Therefore, such modifications or improvements, which do not depart from the spirit of the present application, are within the scope of protection claimed in the present application.

Claims

1. A motor temperature monitoring and control method based on intelligent control, characterized in that: include: Performing a data acquisition cycle specifically includes: determining whether the sampling period has been reached; if so, reading the multi-mode sensor data and outputting the calibrated temperature value T; if not, eliminating vibration noise, compensating for the ambient temperature, and appending the multi-mode sensor data before outputting the calibrated temperature value T; The calibrated temperature value T obtained in the data acquisition cycle is intelligently analyzed: if the temperature is normal, data analysis and prediction are performed; if the temperature is abnormal, the calibrated temperature value T is analyzed, fault self-diagnosis is performed, the cause of the abnormality is determined, the control system performs self-adjustment, and a temperature warning is issued.

2. The motor temperature monitoring and control method based on intelligent control according to claim 1 is characterized in that: The data analysis and prediction specifically include: The calibrated temperature value T is uploaded to the data cloud platform, a temperature-time relationship chart is drawn, and the constructed prediction model TTNet is called. The time-temperature relationship sequence in the relationship chart is input. After analysis, the future time-temperature relationship index CHI is output. If the CHI value is abnormal, a preventive maintenance work order is generated and an alarm is sent. If the CHI value is normal, the data collection loop returns.

3. The motor temperature monitoring and control method based on intelligent control according to claim 1 is characterized in that: After analyzing the calibrated temperature value T, the method further includes: Calculate the temperature rise rate. If the temperature rise is too fast, the control system will self-adjust, generate a maintenance report, and send an alarm. If the temperature rise rate reaches the shutdown threshold, an emergency shutdown will be initiated, power will be cut off, an alarm will be sounded, pre-fault data will be saved, and a fault analysis report will be generated.

4. The motor temperature monitoring and control method based on intelligent control according to claim 1 is characterized in that: The temperature warning specifically includes: By analyzing environmental parameters such as ambient temperature and relative humidity, a dynamic mapping relationship between environmental parameters and warning thresholds is established, and the warning temperature is adjusted in real time. When the temperature reaches the general temperature warning, the control system self-adjusts, generates a maintenance report, and sends an alarm. When the temperature reaches the emergency shutdown threshold, the power is cut off, an alarm is triggered, pre-fault data is saved, and a fault analysis report is generated.

5. The motor temperature monitoring and control method based on intelligent control according to claim 1 is characterized in that: The constructed prediction model TTNet is called, and the time-temperature relationship sequence of the relationship chart is input. After analysis, the future time-temperature relationship index CHI is output, which specifically includes: The input data of the prediction model TTNet includes: Time series features: including time and temperature relationship sequence, Auxiliary features: including vibration signal spectrum, motor current load, ambient temperature and humidity; The output of the prediction model TTNet includes: The predicted temperature and temperature rise rate for the next m time steps.

6. The motor temperature monitoring and control method based on intelligent control according to claim 5 is characterized in that: The network architecture of the constructed prediction model TTNet includes: The spatiotemporal feature encoding layer is used to fuse multimodal input data and extract spatiotemporal correlation features; Multi-scale temporal attention module for dynamically weighting features at different time scales; Thermodynamic residual constraint layer, used to constrain network output through physical equations; Online adaptive module, used to dynamically update model parameters based on real-time data; The fusing of multimodal input data and extracting spatiotemporal correlation features specifically includes: Temporal Convolutional Network: Uses dilated causal convolution to capture multi-scale temporal dependencies and output feature maps. Thermodynamic physics encoder: Input: time-temperature relationship sequence and ambient temperature, calculate heat conduction residual, and embed the residual as a constraint term into the feature vector; The dynamic weighting of different time-scale features specifically includes: calculating attention weights for the multi-scale features output by the temporal convolutional network, and outputting fused features; The output of the network is constrained by physical equations, specifically including: Define the loss function to include data error and physical residual, and control the strength of physical constraints through hyperparameters; The dynamic updating of model parameters according to real-time data specifically includes: When the prediction error of multiple consecutive time steps exceeds the threshold, incremental learning is triggered to use elastic weight solidification to prevent catastrophic forgetting.

7. The motor temperature monitoring and control method based on intelligent control according to claim 6 is characterized in that: The triggering of incremental learning specifically includes: Error monitoring: Real-time calculation of the mean absolute error between the predicted temperature and the measured temperature. If the mean absolute error for N consecutive time steps exceeds the threshold, incremental learning is triggered. Data buffer pool: stores the temperature, vibration, and current time series data of the last K hours. It uses a ring buffer strategy, with new data overwriting old data, and the memory usage remains constant. The elastic weight solidification method specifically includes: Introduce a regularization term into the loss function: ， in: : prediction loss of new data (MSE); : Model parameters before incremental learning; : Fisher information matrix diagonal elements, quantization parameters the importance of : Regularization strength.

8. The motor temperature monitoring and control method based on intelligent control according to claim 1 is characterized in that: The fault self-diagnosis specifically includes: Fault knowledge base construction: By encoding typical temperature-related faults of motors, recording fault types and typical temperature rise characteristics; labeling historical fault cases to build a training set; Fault self-diagnosis logic: Online Fault Detection: Anomaly Scores calculate: ， in, : Historical baseline value of temperature standard deviation under current working conditions, 、 : weight coefficient, : Actual temperature value, : Temperature prediction value, : Normalized value of temperature rise rate, Trigger condition: If If it is greater than the set value, it is determined to be a potential fault and the root cause analysis is started. Fault pattern matching: extract key feature vectors from the output, calculate the similarity between the feature vectors and the historical fault features in the knowledge base, and return the corresponding fault code if the similarity is greater than the set value; otherwise, mark it as "unknown fault". Unknown fault learning: Online clustering and cluster analysis are performed on "unknown fault" samples. If the number of samples in the new category is greater than 50, a temporary fault code is created and the data related to the temporary fault code is pushed to the operation and maintenance personnel, who then request labeling and add it to the knowledge base.

9. A motor temperature monitoring and control system based on intelligent control that executes the method according to any one of claims 1 to 8, characterized in that: include: Adjustment mechanism, multimodal sensing device, intelligent control system, data cloud platform, among which, The multi-modal sensing device uploads the collected multi-modal sensing data such as motor temperature, environmental parameters, vibration spectrum, etc. to the intelligent control system; The intelligent control system reads the sensor data and performs intelligent analysis, and then uploads compressed temperature history data, control logs, fault events and other data to the data cloud platform; The data cloud platform is deployed with a prediction model TTNet for predicting future temperature trends and calculating the future time-temperature relationship index CHI; The data cloud platform transmits cycle optimization suggestions, load adjustment strategies, and model parameter updates to the intelligent control system; The intelligent control system updates various coefficients and adjusts the parameters of the motor body and the regulating mechanism; The intelligent control system transmits control instructions to the motor body and the regulating mechanism, adjusts the motor state in real time, and maintains motor safety and temperature stability.

10. A computer storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a machine, the steps of the method according to any one of claims 1 to 8 are implemented.

Images