Equipment state model training and monitoring method and system, equipment and medium

Abstract

The invention provides an equipment state model training and monitoring method and system, equipment and a medium. The training method comprises the following steps: acquiring first noise data in a target area; wherein target equipment is arranged in the target area; acquiring operation data and state parameters of target equipment with the same time sequence as the first noise data; and taking the first noise data and the state parameter as first training data, taking the operation data as second training data, inputting the first training data and the second training data into a preset neural network, outputting an evaluation state of the target equipment, and training to obtain a target equipment state model. By monitoring noise data of a device group, combining corresponding device operation data and state parameters and based on a machine learning algorithm, a device state model is trained in real time, the device state is monitored, when the device state model deviates from a health model, early warning is triggered, rapid system fault positioning can be assisted, the fault repairing time can be shortened, and the fault repairing efficiency is improved. Therefore, major equipment faults and equipment accidents are avoided.

Description

Technical Field

[0001] The present disclosure relates to the technical field of equipment monitoring, and in particular to a training and monitoring method, system, equipment and medium for an equipment status model. Background Art

[0002] To promote decision-making and deployment for the development of artificial intelligence, coordinate the promotion of artificial intelligence scenario innovation, focus on solving major application and industrialization problems of artificial intelligence, comprehensively improve the quality and level of artificial intelligence development, and better support high-quality development.

[0003] Guided by the creative application of new technologies and taking the linkage of supply and demand as the path, the process of iterative upgrading of new technologies and rapid growth of the industry is realized. Promoting innovation in AI scenarios is of great significance for promoting higher-level applications of AI and better supporting high-quality development. With the rapid development of AI technology, the increasing abundance of data and computing resources, and the continuous expansion of application scenarios, a solid foundation has been laid for the development of AI scenario innovation. However, there are still problems such as insufficient understanding of scenario innovation, insufficient design of major scenario systems, insufficient openness of scenario opportunities, and imperfect scenario innovation ecology. It is necessary to strengthen the overall guidance of AI scenario innovation.

[0004] The digitalization and intelligentization of the energy industry is gradually advancing, and digital technology is deeply integrated with the development of the energy industry, tapping the value of data in traditional industries and ensuring and improving the stability and security of the energy industry.

[0005] The operation of the main equipment of a power plant is inseparable from the coordinated operation of auxiliary equipment such as coal mills, induced draft fans, forced draft fans, feed water pumps, main oil pumps, primary water pumps, circulating water pumps, and condensate pumps. Most of the auxiliary equipment has few measurement parameters on site, the number of auxiliary equipment is large, and some equipment does not have the conditions for adding sensor measurements later. However, the stable operation of auxiliary equipment is directly related to the stable operation of the main equipment. Regular monitoring can only be carried out by manual inspections, which makes it difficult to locate faults in a timely manner. The fault monitoring of equipment is unstable and cannot meet the monitoring purpose of avoiding major equipment failures and equipment accidents. Summary of the invention

[0006] The technical problem to be solved by the present disclosure is to overcome the defects in the prior art that it is difficult to locate faults in a timely manner and the fault monitoring of equipment is unstable, and to provide a training and monitoring method, system, equipment and medium for an equipment status model.

[0007] The present invention solves the above technical problems through the following technical solutions:

[0008] In a first aspect, a method for training a device state model is provided, the training method comprising:

[0009] Acquire first noise data in the target area;

[0010] Wherein, a target device is arranged in the target area;

[0011] Acquire operation data and state parameters of the target device having the same timing as the first noise data;

[0012] The first noise data and the state parameter are used as first training data, and the operation data is used as second training data. The first training data and the second training data are input into a preset neural network, and the evaluation state of the target device is output, so as to obtain a target device state model through training.

[0013] Preferably, the first noise data includes at least one of a noise power spectrum, a noise characteristic frequency and a noise characteristic amplitude.

[0014] Preferably, the step of inputting the first training data and the second training data into a preset neural network, outputting the state evaluation result of the target device, and training to obtain the target device state model comprises:

[0015] Acquire the evaluation status corresponding to the target devices of different types;

[0016] Normalizing the different evaluation states to obtain a target evaluation state;

[0017] The target evaluation state is used as output, the preset neural network is adjusted, and the target device state model is obtained through training.

[0018] Preferably, the target equipment includes at least one of a coal mill, an induced draft fan, a forced draft fan, a feed water pump, a main oil pump, a primary water pump, a circulating water pump and a condensate pump.

[0019] In a second aspect, a method for monitoring a device status is provided, the monitoring method comprising:

[0020] Acquiring second noise data of the area to be monitored;

[0021] Wherein, the equipment to be monitored is arranged in the area to be monitored;

[0022] Acquiring real-time operating data of the equipment to be monitored;

[0023] Inputting the second noise data and the real-time operation data into a target device state model to obtain a state evaluation result of the device to be monitored;

[0024] Wherein, the target device state model is trained according to the device state model training method described in the first aspect.

[0025] In a third aspect, a training system for a device state model is provided, the training system comprising a first noise data acquisition module, a device data acquisition module and a model training module;

[0026] The noise data acquisition module is used to acquire first noise data in a target area; wherein a target device is arranged in the target area;

[0027] The device data acquisition module is used to acquire the operating data and state parameters of the target device with the same time sequence as the first noise data;

[0028] The model training module is used to use the first noise data and the state parameter as the first training data, and the operation data as the second training data, input the first training data and the second training data into a preset neural network, output the evaluation state of the target device, and train to obtain a target device state model.

[0029] In a fourth aspect, a monitoring system for equipment status is provided, the monitoring system comprising a second noise data acquisition module, an operation data acquisition module and an equipment status evaluation module;

[0030] The second noise data acquisition module is used to acquire second noise data of the area to be monitored; wherein the equipment to be monitored is arranged in the area to be monitored;

[0031] The operation data acquisition module is used to acquire the real-time operation data of the device to be monitored;

[0032] The equipment status evaluation module is used to input the second noise data and the real-time operation data into a target equipment status model to obtain a status evaluation result of the equipment to be monitored;

[0033] Wherein, the target device state model is trained according to the training system of the device state model described in the third aspect.

[0034] In a fifth aspect, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and used to run on the processor, wherein when the processor executes the computer program, the method for training the device state model as described in the first aspect and the method for monitoring the device state as described in the second aspect are implemented.

[0035] In a sixth aspect, a computer storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the training method of the device status model as described in the first aspect and the monitoring method of the device status as described in the second aspect are implemented.

[0036] In a seventh aspect, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the method for training the device state model as described in the first aspect, and the method for monitoring the device state as described in the second aspect.

[0037] On the basis of being in accordance with the common sense in the art, the above-mentioned preferred conditions can be arbitrarily combined to obtain the preferred embodiments of the present disclosure.

[0038] The positive progressive effect of the present disclosure is that by monitoring the noise data of the equipment group, combining the corresponding equipment operation data and status parameters, the equipment status model is trained in real time based on the machine learning algorithm, and the equipment status is monitored. When the equipment status model deviates from the healthy model, an early warning is triggered, which can assist in quickly locating system faults, shorten the fault repair time, and thus avoid major equipment failures and equipment accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 A flowchart of a method for training a device state model provided by an exemplary embodiment of the present disclosure;

[0040] Figure 2 A flowchart of a method for monitoring a device status provided by an exemplary embodiment of the present disclosure;

[0041] Figure 3 A schematic diagram of modules of a training system for a device state model provided by an exemplary embodiment of the present disclosure;

[0042] Figure 4 A schematic diagram of a module of a monitoring system for device status provided by an exemplary embodiment of the present disclosure;

[0043] Figure 5 A schematic diagram of the hardware structure of an electronic device provided by an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION

[0044] The present disclosure is further described below by way of examples, but the present disclosure is not limited to the scope of the examples.

[0045] Example 1

[0046] This embodiment provides a method for training a device state model. Figure 1 As shown, the training method comprises:

[0047] S101, acquiring first noise data in a target area;

[0048] Wherein, a target device is arranged in the target area;

[0049] S102, acquiring operation data and state parameters of the target device having the same timing as the first noise data;

[0050] S103, using the first noise data and the state parameter as first training data, using the operation data as second training data, inputting the first training data and the second training data into a preset neural network, outputting the evaluation state of the target device, and training to obtain a target device state model.

[0051] In step S101, a target area where the target device group is located is selected, a noise signal sensor is set in the target area, and a noise signal processing device collects and processes the noise signal sent by the noise signal sensor to obtain the first noise data. At the same time, the power spectrum, characteristic frequency, characteristic amplitude and other data of the collected noise are cleaned to remove abnormal values ​​and erroneous data that may exist therein. For example, if the data at a certain moment deviates significantly from the normal range due to sensor failure or external interference during the process of collecting the noise power spectrum, it needs to be removed. The first noise data is standardized so that parameters of different magnitudes can be compared and analyzed on the same scale. For example, the characteristic amplitude is normalized so that it is within a certain numerical range to facilitate the processing of subsequent models. Among them, the training data can be historical data or real-time collected data.

[0052] In step S102, the corresponding operating data and status parameters at the same time sequence as the first noise data are obtained. The same time sequence may be the operating data and status parameters corresponding to the same collection moment, the same collection cycle or the same collection time period when the target device collects the first noise data. The operating data may include the three-phase voltage and current of the motor, the pump / fan outlet flow rate, pressure, etc., and the status parameters may include normal operation, shutdown or failure, etc.

[0053] In step S103, the training data used to train the preset neural network is divided into first training data and second training data, the first training data includes first noise data and state parameters of the target device in the same time period, and the second training data includes operation data of the target device in the same time period. The first training data and the second training data are used as input variables respectively, and different categories of the overall operation status of the device are determined and labeled according to the actual operation status and business scenario of the target device. Specifically, different types of target devices are processed to obtain corresponding health levels based on corresponding working principles, and the state parameters of the target devices are labeled with health levels. The overall operation status of these labeled devices is normalized and used as the output variable of the model.

[0054] According to the target devices and application scenarios in the target area, select appropriate models, such as convolutional neural networks, decision tree models, support vector machine models, artificial neural network models, etc., and use a large amount of labeled training data to train the selected models. During the training process, the model continuously adjusts its own parameters by learning the relationship between input variables and output variables to achieve accurate evaluation of the overall operating status of the target devices in the target area.

[0055] In this solution, by monitoring the noise data of the equipment group, combining the corresponding equipment operation data and status parameters, the equipment status model is trained in real time based on the machine learning algorithm to monitor the equipment status. When the equipment status model deviates from the health threshold, an early warning is triggered, which can assist in quickly locating system faults, shorten the fault repair time, and thus avoid major equipment failures and equipment accidents.

[0056] As an achievable manner, the trained target device state model is retrained in response to meeting a retraining condition;

[0057] Among them, the retraining conditions include deployment time, the number of abnormal status of the target device, etc.

[0058] In this solution, when the deployment time of the target device state model is greater than the retraining time threshold, the target device state model is retrained. Specifically, the target device state model is retrained using the deployment period operation data, deployment period operation noise data, and corresponding deployment period state parameters corresponding to the target device within the deployment time before retraining as training data, and the parameters of the target device state model are adjusted to obtain a new target device state model, which is then deployed.

[0059] Taking a convolutional neural network as an example, the parameters of the target device state model that are retrained and adjusted include at least one of convolutional layer parameters, pooling layer parameters, fully connected layer parameters, algorithm optimization parameters, and regularization parameters.

[0060] As an implementable manner, the first noise data includes at least one of a noise power spectrum, a noise characteristic frequency, and a noise characteristic amplitude.

[0061] In this scheme, the power spectrum, characteristic frequency and characteristic amplitude of the noise can reflect the noise characteristics generated during the operation of the target equipment, and the target equipment group in the target area maintains the same operating state during long-term operation, and the corresponding noise characteristics change stably. When the operating state of the target equipment in the target area changes, different noise characteristics may be manifested. For example, when the equipment fails, it may cause the vibration of mechanical parts to intensify, thereby changing the power spectrum of the noise, and the characteristic frequency and characteristic amplitude may also become abnormal. These noise-related parameters are combined with the state parameters of the equipment under test in the same time period as input variables because the state of the equipment itself (such as temperature, speed and other state parameters) will directly affect the generation of noise. At the same time, the change in noise can also reflect the operating state of the equipment in turn. The two are interrelated and can more comprehensively characterize the operation of the equipment at a certain moment.

[0062] As an achievable manner, the step of inputting the first training data and the second training data into a preset neural network, outputting the state evaluation result of the target device, and training to obtain the target device state model includes:

[0063] Acquire the evaluation status corresponding to the target devices of different types;

[0064] Normalizing the different evaluation states to obtain a target evaluation state;

[0065] The target evaluation state is used as output, the preset neural network is adjusted, and the target device state model is obtained through training.

[0066] In this solution, in actual business scenarios, the overall operating status of the evaluated equipment is normalized as the output variable of the model, mainly to facilitate comparison and analysis between different equipment and different time periods of the same equipment. In power plants, the operating status of fans and pumps may be expressed in multiple ways. For example, fans may use specific numerical ranges of air volume and air pressure to indicate whether they are normal or not, and pumps may use specific values ​​of flow rate and head to define the operating status. Through normalization, these different forms of evaluation results can be converted into unified values ​​within a certain range. For example, the normal operating status is normalized to a value between 0 and 1 (such as 0.8 - 1 for normal), the minor fault status may be normalized to a value between 0.5 - 0.8, and the serious fault status is normalized to a value between 0 and 0.5 (the specific value can be adjusted according to the actual situation and subsequent analysis requirements). Regardless of different fans and pumps, or the operating status of the same equipment at different times, they can be intuitively compared and judged on the same scale, which is convenient for long-term monitoring and management of the operating status of power plant equipment.

[0067] As an achievable manner, the target equipment includes at least one of a coal mill, an induced draft fan, a forced draft fan, a feed water pump, a main oil pump, a primary water pump, a circulating water pump and a condensate pump.

[0068] In this scheme, this embodiment is applied to fans and / or pump equipment in power plants, and important driving equipment in the air, water and / or oil systems of the generator sets are used as target equipment. By monitoring key target equipment, major equipment failures and equipment accidents can be avoided.

[0069] The training method of the equipment status model provided in this embodiment monitors the noise data of the equipment group, combines the corresponding equipment operation data and status parameters, and trains the equipment status model in real time based on a machine learning algorithm to monitor the equipment status. When the equipment status model deviates from the healthy model, an early warning is triggered, which can assist in quickly locating system faults, shorten the fault repair time, and thus avoid major equipment failures and equipment accidents.

[0070] Example 2

[0071] This embodiment provides a method for monitoring the status of a device, such as Figure 2 As shown, the monitoring method comprises:

[0072] S201, obtaining second noise data of the area to be monitored;

[0073] Wherein, the equipment to be monitored is arranged in the area to be monitored;

[0074] S202, obtaining real-time operation data of the device to be monitored;

[0075] S203, inputting the second noise data and the real-time operation data into a target device state model to obtain a state evaluation result of the device to be monitored;

[0076] The target device state model is trained according to the device state model training method described in Example 1.

[0077] In this solution, the area where the equipment to be monitored is located is selected as the area to be monitored, and a number of noise signal sensors for obtaining the second noise data are installed in the area to be monitored. The second noise data is collected and processed through the noise signal collection and processing device, and the data is sent to the equipment status monitoring system through a specific transmission protocol. At the same time, the measurement data of other systems of the equipment to be monitored are connected as the corresponding actual operation data. With the second noise data and real-time operation data as input, the real-time status evaluation of the equipment to be monitored is obtained through the target equipment status model. When the equipment status model deviates from the health model, an early warning is triggered, which can assist in the rapid location of system faults, shorten the fault repair time, and thus avoid major equipment failures and equipment accidents.

[0078] The device status monitoring method provided in this embodiment obtains the second noise data and real-time operation data of the device to be monitored as the input data of the model, and realizes status monitoring of the device to be monitored based on the target device status model. When the status of the device to be monitored deviates from the healthy model, an early warning is triggered to assist in the rapid location of the fault, which can shorten the fault repair time and avoid major equipment failures and equipment accidents.

[0079] Example 3

[0080] This embodiment provides a training system 100 for a device state model, such as Figure 3 As shown, the training system 100 includes a first noise data acquisition module 101, a device data acquisition module 102 and a model training module 103;

[0081] The first noise data acquisition module 101 is used to acquire first noise data in a target area; wherein a target device is arranged in the target area;

[0082] The device data acquisition module 102 is used to acquire the operating data and state parameters of the target device with the same time sequence as the first noise data;

[0083] The model training module 103 is used to use the first noise data and the state parameter as the first training data, and the operation data as the second training data, input the first training data and the second training data into a preset neural network, output the evaluation state of the target device, and train to obtain a target device state model.

[0084] As an implementable manner, the first noise data includes at least one of a noise power spectrum, a noise characteristic frequency, and a noise characteristic amplitude.

[0085] As an implementable manner, the model training module 103 includes an evaluation state acquisition unit, a normalization processing unit and an adjustment unit;

[0086] An evaluation status acquisition unit, used to acquire the evaluation status corresponding to different types of target devices;

[0087] A normalization processing unit, used for normalizing the different evaluation states to obtain a target evaluation state;

[0088] An adjustment unit is used to adjust the preset neural network with the target evaluation state as output, and train to obtain the target device state model.

[0089] As an achievable manner, the target equipment includes at least one of a coal mill, an induced draft fan, a forced draft fan, a feed water pump, a main oil pump, a primary water pump, a circulating water pump and a condensate pump.

[0090] As for the system embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The system embodiment described above is only illustrative, wherein the units described as separate components may or may not be physically separated, and the components as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the disclosed solution.

[0091] The equipment status model training system provided in this embodiment monitors the noise data of the equipment group, combines the corresponding equipment operation data and status parameters, and trains the equipment status model in real time based on a machine learning algorithm to monitor the equipment status. When the equipment status model deviates from the healthy model, an early warning is triggered, which can assist in quickly locating system faults, shorten fault repair time, and thus avoid major equipment failures and equipment accidents.

[0092] Example 4

[0093] This embodiment provides a device status monitoring system 200, such as Figure 4 As shown, the monitoring system 200 includes a second noise data acquisition module 201, an operation data acquisition module 202 and an equipment status evaluation module 203;

[0094] The second noise data acquisition module 201 is used to acquire second noise data of the area to be monitored; wherein the equipment to be monitored is arranged in the area to be monitored;

[0095] The operation data acquisition module 202 is used to acquire the real-time operation data of the device to be monitored;

[0096] The equipment status evaluation module 203 is used to input the second noise data and the real-time operation data into a target equipment status model to obtain a status evaluation result of the equipment to be monitored;

[0097] The target device state model is obtained by training according to the device state model training system described in Example 3.

[0098] The equipment status monitoring system provided in this embodiment obtains the second noise data and real-time operation data of the equipment to be monitored as the input data of the model, and realizes status monitoring of the equipment to be monitored based on the target equipment status model. When the status of the equipment to be monitored deviates from the healthy model, an early warning is triggered to assist in the rapid location of the fault, which can shorten the fault repair time and thus avoid the occurrence of major equipment failures and equipment accidents.

[0099] Example 5

[0100] Figure 5 This is a structural diagram of an electronic device shown in an example embodiment of the present disclosure, the electronic device includes a memory, a processor, and a computer program stored in the memory and used to run on the processor, and when the processor executes the computer program, it implements the training method of the device state model or the monitoring method of the device state described in any of the above embodiments. Figure 5 The electronic device 90 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0101] like Figure 5 As shown, the electronic device 90 may be in the form of a general-purpose computing device, for example, it may be a server device. The components of the electronic device 90 may include, but are not limited to: at least one processor 91, at least one memory 92, and a bus 93 connecting different system components (including the memory 92 and the processor 91).

[0102] The bus 93 includes a data bus, an address bus, and a control bus.

[0103] The memory 92 may include a volatile memory, such as a random access memory (RAM) 921 and / or a cache memory 922 , and may further include a read only memory (ROM) 923 .

[0104] The memory 92 may also include a program tool 925 (or utility) having a set (at least one) of program modules 924, such program modules 924 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0105] The processor 91 executes various functional applications and data processing by running the computer program stored in the memory 92, such as the device status model training method or device status monitoring method described in any of the above embodiments.

[0106] The electronic device 90 may also communicate with one or more external devices 94 (e.g., keyboards, pointing devices, etc.). Such communication may be performed via an input / output (I / O) interface 95. Furthermore, the electronic device 90 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 96. As shown, the network adapter 96 communicates with other modules of the electronic device 90 via a bus 93. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 90, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems, etc.

[0107] It should be noted that although several units / modules or sub-units / modules of the electronic device are mentioned in the above detailed description, this division is merely exemplary and not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided into multiple units / modules to be embodied.

[0108] Example 6

[0109] The embodiments of the present disclosure also provide a computer-readable storage medium having a computer program stored thereon, and when the program is executed by a processor, the method for training a device state model or the method for monitoring a device state described in any of the above embodiments is implemented.

[0110] The readable storage medium may include but is not limited to: a portable disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical storage device, a magnetic storage device or any suitable combination of the above.

[0111] Example 7

[0112] The embodiments of the present disclosure also provide a computer program product, including a computer program, which, when executed by a processor, implements any of the above-mentioned methods for training a device state model or monitoring a device state.

[0113] Among them, the program code for executing the computer program product of the present disclosure can be written in any combination of one or more programming languages, and the program code can be executed completely on the user device, partially on the user device, as an independent software package, partially on the user device and partially on a remote device, or completely on the remote device.

[0114] Although the specific embodiments of the present disclosure are described above, those skilled in the art should understand that this is only an example, and the protection scope of the present disclosure is defined by the appended claims. Those skilled in the art may make various changes or modifications to these embodiments without departing from the principles and essence of the present disclosure, but these changes and modifications all fall within the protection scope of the present disclosure.

Claims

1. A method for training a device state model, characterized in that: The training method comprises: Acquire first noise data in the target area; Wherein, a target device is arranged in the target area; Acquire operation data and state parameters of the target device having the same timing as the first noise data; The first noise data and the state parameter are used as first training data, and the operation data is used as second training data. The first training data and the second training data are input into a preset neural network, and the evaluation state of the target device is output, so as to obtain a target device state model through training.

2. The method for training a device state model according to claim 1, characterized in that: The first noise data includes at least one of a noise power spectrum, a noise characteristic frequency, and a noise characteristic amplitude.

3. The method for training a device state model according to claim 1, characterized in that: The step of inputting the first training data and the second training data into a preset neural network, outputting the state evaluation result of the target device, and training to obtain the target device state model comprises: Acquire the evaluation status corresponding to the target devices of different types; Normalizing the different evaluation states to obtain a target evaluation state; The target evaluation state is used as output, the preset neural network is adjusted, and the target device state model is obtained through training.

4. The method for training a device state model according to claim 1, characterized in that: The target equipment includes at least one of a coal mill, an induced draft fan, a forced draft fan, a feed water pump, a main oil pump, a primary water pump, a circulating water pump and a condensate pump.

5. A method for monitoring equipment status, characterized in that: The monitoring method comprises: Acquiring second noise data of the area to be monitored; Wherein, the equipment to be monitored is arranged in the area to be monitored; Acquiring real-time operating data of the equipment to be monitored; Inputting the second noise data and the real-time operation data into a target device state model to obtain a state evaluation result of the device to be monitored; The target device state model is trained according to the device state model training method according to any one of claims 1 to 4.

6. A training system for a device state model, characterized in that: The training system includes a first noise data acquisition module, a device data acquisition module and a model training module; The noise data acquisition module is used to acquire first noise data in a target area; wherein a target device is arranged in the target area; The device data acquisition module is used to acquire the operating data and state parameters of the target device with the same time sequence as the first noise data; The model training module is used to use the first noise data and the state parameter as the first training data, and the operation data as the second training data, input the first training data and the second training data into a preset neural network, output the evaluation state of the target device, and train to obtain a target device state model.

7. A monitoring system for equipment status, characterized in that: The monitoring system includes a second noise data acquisition module, an operation data acquisition module and an equipment status assessment module; The second noise data acquisition module is used to acquire second noise data of the area to be monitored; wherein the equipment to be monitored is arranged in the area to be monitored; The operation data acquisition module is used to acquire the real-time operation data of the device to be monitored; The equipment status evaluation module is used to input the second noise data and the real-time operation data into a target equipment status model to obtain a status evaluation result of the equipment to be monitored; Wherein, the target device state model is trained by the device state model training system according to claim 6.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and used to run on the processor, characterized in that: When the processor executes the computer program, it implements the method for training the device status model as described in any one of claims 1 to 4, and the method for monitoring the device status as described in claim 5.

9. A computer storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the training method of the device status model as described in any one of claims 1 to 4 and the monitoring method of the device status as described in claim 5 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, it implements the method for training the device status model as described in any one of claims 1 to 4, and the method for monitoring the device status as described in claim 5.

Images