A device and system for monitoring and diagnosing the status of a lubrication system of a mobile device

Abstract

The present invention relates to a device and system for monitoring and diagnosing the state of a lubrication system of a power-driven device, and relates to the field of power-driven device state monitoring. It includes: after the power-driven device is started, the main collector receives the monitoring status of the lubrication system from the slave collector, calculates the deviation from the theoretical state of the lubrication system, obtains the lubrication system state deviation time series matrix for demand computing power prediction, and obtains the demand computing power prediction value; obtains the edge node task allocation status for edge node scheduling optimization, and obtains the edge node scheduling scheme; activates the target edge node according to the edge node scheduling scheme, and performs device state prediction based on the lubrication system state deviation time series matrix to obtain device state identification information; when including an abnormal state type, it is sent to the management end through the cloud and synchronously displayed with the main collector. It solves the technical problem in the prior art that due to the one-sided monitoring state attributes and limited local computing power, the state of the power-driven device cannot be comprehensively and timely reflected.

Description

Technical Field

[0001] The present invention relates to the field of active equipment status monitoring, and particularly to a device and system for status monitoring and diagnosis of the lubrication system of active equipment. Background Art

[0002] With the development of industrial automation and intelligence, active equipment is increasingly widely used in various fields. The status monitoring and diagnosis of its lubrication system are crucial for ensuring the stable operation of the equipment. The lubrication system is equivalent to the "blood vessels and blood" of the active equipment, and its good operating state is the basis for the normal operation of the equipment, and it can also effectively reflect the operating health status of mechanical equipment.

[0003] There are mainly two modes for the detection of the lubrication system: the offline detection mode of sampling and sending for inspection and the online detection mode of online sampling of lubricating oil: The offline detection mode samples at the sampling points of the lubrication system according to the set sampling and inspection frequency, and then sends it to the enterprise's own or outsourced analysis and inspection unit for detection. However, this mode has low efficiency and poor timeliness, and it is impossible to grasp the status of the lubrication system in real time; Although the online detection mode can monitor the quality and pollutants of lubricating oil in real time and can conduct a preliminary diagnosis on the oil product performance, partial pollution and wear status, it cannot comprehensively reflect the status of active equipment, and because analyzing the working conditions of active equipment based on the status of the lubrication system shows a strong non-linear relationship, the local computing power is insufficient to support the timely analysis of the working conditions of active equipment. Summary of the Invention

[0004] Aiming at the technical problem in the prior art that due to the one-sided monitoring state attributes and limited local computing power, it is impossible to comprehensively and timely reflect the status of active equipment, the present invention provides a device and system for status monitoring and diagnosis of the lubrication system of active equipment to solve this problem.

[0005] The technical solution of the present invention to solve the above technical problems is as follows:

[0006] In a first aspect, the present invention provides a device for monitoring and diagnosing the state of a lubrication system of a power-driven device, and a system for monitoring and diagnosing the state of a lubrication system of a power-driven device is deployed. The system includes a collection end, a cloud end, and a management end. The collection end includes a main collector and a slave collector. The execution steps include: when the power-driven device is started and the timer meets a preset duration, the main collector receives the monitoring state of the lubrication system received by the slave collector, calculates the deviation from the theoretical state of the lubrication system to obtain a time series matrix of the lubrication system state deviation, and at the same time resets the timer to zero and starts timing again. Among them, the attributes of the lubrication system monitoring state include flow rate, viscosity, temperature, density, dielectric constant, water activity, water content, metal chip density, particle distribution density, oil-water interface level, oil-gas interface level, pressure, and degree of emulsification; predicting the required computing power according to the time series matrix of the lubrication system state deviation to obtain a predicted value of the required computing power; obtaining the task allocation status of the edge node; performing edge node scheduling optimization according to the predicted value of the required computing power and the task allocation status of the edge node to obtain an edge node scheduling plan; activating a target edge node according to the edge node scheduling plan, and predicting the device state based on the time series matrix of the lubrication system state deviation to obtain device state identification information; when the device state identification information includes an abnormal state type, sending the device state identification information to the management end through the cloud end and performing synchronous display with the main collector.

[0007] In a second aspect, the present invention provides a system for monitoring and diagnosing the state of a lubrication system of a power-driven device, which is deployed in the device for monitoring and diagnosing the state of a lubrication system of a power-driven device in the first aspect.

[0008] The beneficial effects of the present invention are as follows: By setting the flow rate, viscosity, temperature, density, dielectric constant, water activity, water content, metal chip density, particle distribution density, oil-water interface level, oil-gas interface level, pressure, and degree of emulsification as monitoring attributes, the state of the lubrication system can be comprehensively monitored, improving the comprehensiveness of the monitoring state, thus solving the problem that the prior art cannot overall reflect the lubrication state of the device lubrication system. At the same time, using the time series matrix of the lubrication system state deviation to predict the required computing power and performing edge node scheduling to execute device state analysis based on the predicted value of the required computing power can reasonably allocate tasks and optimize the use of computing power according to limited local computing power resources, solve the problem of insufficient local computing power, achieve the technical effect of efficiently predicting the state of the power-driven device according to the state of the lubrication system, provide strong technical support for the fault diagnosis and preventive maintenance of the power-driven device, and improve the reliability and stability of the device operation. Description of the Drawings

[0009] Figure 1 It is a schematic flow chart of the execution steps of a device for monitoring and diagnosing the state of a lubrication system of a power-driven device provided by the present invention;

[0010] Figure 2 Schematic structural diagram of a system for monitoring and diagnosing the lubrication system status of active equipment provided by the present invention;

[0011] Figure 3 Possible application deployment diagram of a device for monitoring and diagnosing the lubrication system status of active equipment provided by the present invention. Specific embodiments

[0012] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present invention.

[0013] In the description of the present invention, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of the described features. In the description of the present invention, "a plurality" means two or more, unless otherwise specifically defined.

[0014] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration, or explanation". Any embodiment described as "for example" in the present invention is not necessarily construed as being more preferred or having more advantages than other embodiments. The following description is provided to enable any person skilled in the art to implement and use the present invention. In the following description, details are set forth for purposes of explanation. It should be understood that those skilled in the art can recognize that the present invention can be implemented without using these specific details. In other instances, well-known structures and processes are not described in detail to avoid unnecessary details from obscuring the description of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.

[0015] Embodiment 1:

[0016] As Figure 1 shown, the embodiment of the present invention provides a device for monitoring and diagnosing the lubrication system status of active equipment, deployed with a system for monitoring and diagnosing the lubrication system status of active equipment. The system includes a collection end, a cloud end, and a management end. The collection end includes a main collector and a slave collector. The execution steps include the following steps:

[0017] Specifically, the system for monitoring and diagnosing the status of the lubrication system of active equipment is software, and the device for monitoring and diagnosing the status of the lubrication system of active equipment is hardware, such as a storage medium. The acquisition end is a data acquisition interface for communicating with the lubrication system sensor and acquiring the status data of the lubrication system. The cloud is a virtual medium for remote data interaction. The management end is a terminal for remotely managing active equipment. Preferably, it includes a PC terminal, a mobile terminal, and a control room DCS system, etc. The main collector is a functional module for receiving and storing the data of the slave collector, and the slave collector is a functional module for directly receiving signals from the lubrication system sensor. Preferably, generally a set of "master and slave" collectors serves one active equipment. If necessary, multiple slave collectors can also be set on different devices, and one main collector is connected to these slave collectors, so that a set of "master and slave" collectors can also serve multiple devices. The master and slave collectors can communicate either wired or wirelessly. The display is divided into on-site display of the main collector, display on the management end, that is, display on the control room DCS system, display on the PC terminal, and display on the mobile terminal, corresponding to on-site, wired communication, wireless communication, or the cloud platform respectively.

[0018] Further, as Figure 3 shown in the deployment schematic diagram of the possible device for monitoring and diagnosing the status of the lubrication system of active equipment, the on-site collector is the "master and slave" collector, DCS is the control room DCS system, the server is the cloud, the management end, and the application service is the mobile terminal.

[0019] S10: After the active equipment is started and the timer meets the preset duration, the main collector receives the monitoring status of the lubrication system received by the slave collector, calculates the deviation from the theoretical status of the lubrication system to obtain the lubrication system status deviation time series matrix, and at the same time resets the timer to zero and starts timing again. Among them, the attributes of the lubrication system monitoring status include flow rate, viscosity, temperature, density, dielectric constant, water activity, water content, metal chip density, particle distribution density, oil-water interface level, oil-gas interface level, pressure, and degree of emulsification.

[0020] Specifically, the start of the active equipment means that the active equipment starts to run. The timer meeting the preset duration refers to a preset timing mechanism. When the preset time threshold is reached, subsequent acquisition and calculation operations are triggered, and the timer is reset to zero and starts timing again every time after acquisition. The main collector is responsible for receiving the lubrication system monitoring status data sent by the slave collector. Preferably, the slave collector is configured according to the data attributes to be monitored, and at least includes sensors corresponding to flow rate, viscosity, temperature, density, dielectric constant, water activity, water content, metal chip density, particle distribution density, oil-water interface level, oil-gas interface level, pressure, and degree of emulsification. The theoretical status of the lubrication system refers to the threshold value of the lubrication system monitoring status pre-configured by the user each time the active equipment is started.

[0021] Taking the theoretical state of the lubrication system as the minuend and the monitored state of the lubrication system as the subtrahend, calculate the difference of the same attributes at each moment between the monitored state of the lubrication system and the theoretical state of the lubrication system to obtain the deviation vector of each attribute at each moment; further, taking the moment as the horizontal element of the matrix and the attribute as the vertical element of the matrix, store the deviation vector of each attribute at each moment to obtain the lubrication system state deviation time series matrix for reflecting the change of the lubrication system state over time.

[0022] S20: Perform demand computing power prediction according to the lubrication system state deviation time series matrix to obtain the demand computing power prediction value;

[0023] Further, perform demand computing power prediction according to the lubrication system state deviation time series matrix to obtain the demand computing power prediction value. Step S20 includes:

[0024] Step S21: Perform fuzzy deviation parameter analysis according to the lubrication system state deviation time series matrix to obtain the first monitored attribute fuzzy deviation parameter to the Nth monitored attribute fuzzy deviation parameter;

[0025] Step S22: According to the first monitored attribute fuzzy deviation parameter to the Nth monitored attribute fuzzy deviation parameter, extract the set of monitored attributes whose fuzzy deviation parameters are greater than or equal to the fuzzy deviation threshold;

[0026] Step S23: Perform demand computing power prediction according to the set of monitored attributes to obtain the demand computing power prediction value.

[0027] Further, perform demand computing power prediction according to the set of monitored attributes to obtain the demand computing power prediction value. Step S23 includes the steps:

[0028] S231: Obtain the device state prediction unit library, where the device state prediction unit library has a number of device state prediction units, and the number of device state prediction units has a number of monitored attribute weight distribution results and a number of demand computing power identifiers;

[0029] S232: Based on the set of monitored attributes, traverse the number of monitored attribute weight distribution results for matching degree analysis to obtain a number of matching degrees;

[0030] S233: According to the number of matching degrees, extract the set of target device state prediction units with matching degrees greater than or equal to the matching degree threshold from the device state prediction unit library;

[0031] S234: According to the set of target device state prediction units, perform demand computing power calculation based on the number of demand computing power identifiers to obtain the demand computing power prediction value.

[0032] Further, the steps of constructing the device state prediction unit library include:

[0033] Match the operation logs of active devices according to the active device models, where the operation logs of active devices include the time series record matrix of the lubrication system state deviation and the record information of the active device state type;

[0034] Obtain the preset weight distribution information of monitoring attributes;

[0035] Configure the input layer of the long short-term memory neural network according to the preset weight distribution information of the monitoring attributes. Using the record information of the active device state type as supervision and the time series record matrix of the lubrication system state deviation as input, train the device state prediction unit and add it to the device state prediction unit library.

[0036] Further, perform fuzzy deviation parameter analysis according to the time series matrix of the lubrication system state deviation to obtain the first monitoring attribute fuzzy deviation parameter to the Nth monitoring attribute fuzzy deviation parameter. Step S21 includes the steps:

[0037] S211: Obtain the set of monitoring attribute deviation thresholds;

[0038] S212: Extract the first monitoring attribute deviation threshold to the Nth monitoring attribute deviation threshold from the set of monitoring attribute deviation thresholds;

[0039] S213: Statistically analyze the proportion of the moments when the deviation vector modulus in the time series information of the first monitoring attribute deviation vector modulus of the time series matrix of the lubrication system state deviation is greater than or equal to the first monitoring attribute deviation threshold, and set it as the first monitoring attribute fuzzy deviation parameter;

[0040] S214: Until statistically analyze the proportion of the moments when the deviation vector modulus in the time series information of the Nth monitoring attribute deviation vector modulus of the time series matrix of the lubrication system state deviation is greater than or equal to the Nth monitoring attribute deviation threshold, and set it as the Nth monitoring attribute fuzzy deviation parameter.

[0041] Further, based on the set of monitoring attributes, traverse the matching degree analysis of the weight distribution results of the several monitoring attributes to obtain several matching degrees. Step S233 includes the steps:

[0042] S2331: Traverse the weight distribution results of the several monitoring attributes, sort the monitoring attributes according to the weights from large to small, and obtain several monitoring attribute sorting results;

[0043] S2332: Obtain the set of monitoring attributes, traverse the several monitoring attribute sequences, and extract several monitoring attribute serial number sequences;

[0044] S2333: Construct a matching degree evaluation function:

[0045]

[0046] where P(X i ) represents the matching degree of the i-th monitoring attribute serial number sequence, n represents the number of monitoring attribute sets, X ij represents the j-th serial number of the i-th monitoring attribute serial number sequence, and X i represents the i-th monitoring attribute serial number sequence;

[0047] S2334: According to the matching degree evaluation function, traverse the several monitoring attribute serial number sequences for matching degree analysis to obtain the several matching degrees.

[0048] Specifically, in the embodiment of the present application, for different lubrication system state deviation time series matrices, since the models that may be scheduled for equipment state prediction are different, the required computing power will also be different. Therefore, each time a lubrication system state deviation time series matrix is obtained, it is necessary to perform demand computing power prediction to obtain a demand computing power prediction value, so as to provide a reference for subsequent task allocation of edge nodes.

[0049] The following is the specific method for realizing demand computing power prediction in the embodiment of the present application:

[0050] The first monitoring attribute fuzzy deviation parameter represents the overall deviation of the first monitoring attribute calculated based on the lubrication system state deviation time series matrix. The detailed calculation method is as follows:

[0051] Through the management terminal, obtain the deviation threshold set of each monitoring attribute, and then extract the deviation threshold of the first monitoring attribute; extract the deviation vector modulus time series information of the first monitoring attribute from the lubrication system state deviation time series matrix; further, calculate the number of moments when the deviation vector modulus in the deviation vector modulus time series information of the first monitoring attribute is greater than or equal to the deviation threshold of the first monitoring attribute, and calculate the ratio of the number of moments to the total number of moments of the deviation vector modulus time series information of the first monitoring attribute, which is set as the first monitoring attribute fuzzy deviation parameter; analyze the second monitoring attribute to the Nth monitoring attribute in the same way, so as to obtain the first monitoring attribute fuzzy deviation parameter to the Nth monitoring attribute fuzzy deviation parameter.

[0052] Further, according to the first monitoring attribute fuzzy deviation parameter to the Nth monitoring attribute fuzzy deviation parameter, extract the set of monitoring attributes whose fuzzy deviation parameters are greater than or equal to the fuzzy deviation threshold. The fuzzy deviation threshold refers to the fuzzy deviation characteristic value preset by the user, and the default value is 0.3. The set of monitoring attributes refers to the monitoring attributes that satisfy the fuzzy deviation parameter being greater than or equal to the fuzzy deviation threshold. When the fuzzy deviation parameter is greater than or equal to the fuzzy deviation threshold, it indicates that the corresponding monitoring attribute has a relatively large overall deviation in this monitoring. Therefore, it needs to be emphasized, so it is stored as the set of monitoring attributes for subsequent calls.

[0053] Furthermore, the device status prediction unit library is used to store the function unit library for device status prediction. Any device status prediction unit is a prediction model based on the long short-term memory neural network under different monitoring attribute weight distributions. Preferably, the training method of any device status prediction unit is as follows:

[0054] Through the management terminal, configure the preset weight distribution information of the monitoring attributes; then match the operating logs of the active devices according to the active device models. Among them, any operating log of the active device includes the lubrication system status deviation time series record matrix and the active device status type record information. The active device status type record information is encoded information, and any active device status type corresponds to a unique code. Use the preset weight distribution information of the monitoring attributes to construct the input layer of the long short-term memory neural network, and perform weighted prediction processing on the lubrication system status deviation time series record matrix.

[0055] Furthermore, divide the operating logs of the active devices into 8 parts of training data sets and 2 parts of validation data sets; at the same time, construct the loss function of the multi-classification model, where, X i0 represents the active device status type record information, X i represents the active device status type prediction information, X i ∩X i0Characterize the intersection of the active device status type record information and the active device status type record information as the active device status type; configure the convergence loss threshold. During specific training, schedule 8 portions of the training data set. Using the active device status type record information as the supervision and the lubrication system status deviation time series record matrix as the input, train the long short-term memory neural network with pre-configured weights. When the LOSS value of continuous preset number of training is less than or equal to the convergence loss threshold, schedule 2 portions of the validation data set. Using the active device status type record information as the supervision and the lubrication system status deviation time series record matrix as the input, train the long short-term memory neural network with pre-configured weights. When at least 50 LOSS values are less than or equal to the convergence loss threshold, consider the long short-term memory neural network to converge, store it as the device status prediction unit, and store the monitoring attribute preset weight distribution information as the monitoring attribute weight distribution result. Further, calculate the average value of the computing power demand values during the operation of the device status prediction unit at least 500 times, and set it as the demand computing power identifier; associate and store the monitoring attribute weight distribution result, the demand computing power identifier with the device status prediction unit into the device status prediction unit library. Preferably, the monitoring attribute weight distribution result will be enumerated and combined. Here, the enumeration and combination refer to the enumeration and combination of the monitoring attributes sorted in descending order of weight, and then the obtained processing unit set can be used to retrieve different device status prediction units for analysis when facing different deviation states to improve the analysis accuracy.

[0056] During specific application, according to the monitoring attribute set, calculate the matching degree of each device status prediction unit in the device status prediction unit library. Preferably, the matching degree calculation method is as follows:

[0057] Traverse the monitoring attribute weight distribution results of several device status prediction units, sort the monitoring attributes in descending order of weight based on each monitoring attribute weight distribution result to obtain several monitoring attribute sorting results; further, traverse several monitoring attribute sorting results, respectively extract the serial number parameters of the monitoring attribute set in each sorting, and store them as several monitoring attribute serial number sequences; furthermore, use the matching degree evaluation function: Process several monitoring attribute serial number sequences respectively to obtain several matching degrees corresponding to several device status prediction units one by one.

[0058] Through the management terminal, configure the matching degree threshold; then according to the several matching degrees, extract the set of target device status prediction units with matching degrees greater than or equal to the matching degree threshold from the device status prediction unit library. At this time, the set of target device status prediction units is the functional component that needs to perform device status prediction in the subsequent steps. Calculate the sum of the demand computing powers according to the several demand computing power identifiers of the target device status prediction unit set to obtain the demand computing power prediction value.

[0059] S30: Obtain the task allocation status of edge nodes;

[0060] S40: Perform edge node scheduling optimization according to the predicted demand computing power value and the task allocation status of edge nodes, and obtain an edge node scheduling plan;

[0061] Further, perform edge node scheduling optimization according to the predicted demand computing power value and the task allocation status of edge nodes, and obtain an edge node scheduling plan. Step S40 includes the steps:

[0062] S41: Obtain the redundant computing power list of the edge node list according to the task allocation status of edge nodes;

[0063] S42: According to the predicted demand computing power value, based on the redundant computing power list, match the target edge nodes and add them to the edge node scheduling plan.

[0064] Specifically, the task allocation status of edge nodes refers to the occupancy status of current edge nodes. Specifically, the task allocation status of edge nodes includes the redundant computing power list of the edge node list at the current moment. Redundant computing power refers to the schedulable idle computing power.

[0065] In one implementation, traverse the redundant computing power list, compare it with the predicted demand computing power value, extract the edge nodes with redundant computing power greater than the predicted demand computing power value, and download the aforementioned target device status prediction unit set to the corresponding edge nodes to obtain an edge node scheduling plan;

[0066] In another implementation, when there are no edge nodes in the redundant computing power list with redundant computing power greater than the predicted demand computing power value, at this time, split the target device status prediction unit set into multiple independent units according to the demand computing power identification set, and then construct a scheduling plan based on the redundant computing power list, and optimize with the number of edge node schedules as the fitness function to obtain an edge node scheduling plan;

[0067] In a third implementation, when the above two methods cannot meet the predicted demand computing power value, generate a prediction warning message and feedback it to the management end.

[0068] S50: Activate the target edge nodes according to the edge node scheduling plan, and perform device status prediction based on the lubrication system state deviation time series matrix to obtain device status identification information;

[0069] S60: When the device status identification information includes an abnormal status type, send the device status identification information to the management end through the cloud and perform synchronous display with the main collector.

[0070] Specifically, according to the edge node scheduling scheme, a target edge node is configured. The target edge node has downloaded a target device state prediction unit for device state prediction. At this time, by inputting the lubrication system state deviation time series matrix into the target device state prediction unit, device state identification information can be output.

[0071] When the device state identification information includes an abnormal state type, for example, when the encodings of abnormal working conditions such as liquid level, interface level change, filter failure, emulsification, etc. are included, the device state identification information is sent to the management end through the cloud, and at the same time, the abnormal state type is synchronously displayed on the main collector, as well as the DCS system, PC end, and mobile end of the management end.

[0072] A device for monitoring and diagnosing the lubrication system state of active devices provided by an embodiment of the present invention has at least the following technical effects:

[0073] By setting flow rate, viscosity, temperature, density, dielectric constant, water activity, water content, metal chip density, particle distribution density, oil-water interface level, oil-gas interface level, pressure, and emulsification degree as monitoring attributes, the state of the lubrication system can be comprehensively monitored, improving the comprehensiveness of the monitored state, thus solving the problem that the prior art cannot overall reflect the lubrication state of the device lubrication system. At the same time, by using the lubrication system state deviation time series matrix for demand computing power prediction and performing device state analysis by edge node scheduling according to the demand computing power prediction value, tasks can be reasonably allocated based on limited local computing power resources, optimizing the use of computing power, solving the problem of insufficient local computing power, achieving the technical effect of efficiently predicting the state of active devices according to the lubrication system state, providing strong technical support for the fault diagnosis and preventive maintenance of active devices, and improving the reliability and stability of device operation.

[0074] Embodiment 2:

[0075] As Figure 2 shown, an embodiment of the present invention further provides a system for monitoring and diagnosing the lubrication system state of active devices, which is deployed on a device for monitoring and diagnosing the lubrication system state of active devices provided in Embodiment 1. The execution steps include:

[0076] After the active device is started and the timer meets the preset duration, the main collector receives the lubrication system monitoring state from the slave collector, calculates the deviation from the theoretical state of the lubrication system to obtain the lubrication system state deviation time series matrix, and at the same time resets the timer to zero and starts timing again, where the lubrication system monitoring state attributes include flow rate, viscosity, temperature, density, dielectric constant, water activity, water content, metal chip density, particle distribution density, oil-water interface level, oil-gas interface level, pressure, and emulsification degree;

[0077] Predict the required computing power according to the timing matrix of the lubrication system state deviation to obtain the predicted value of the required computing power;

[0078] Obtain the task allocation status of the edge node;

[0079] Perform edge node scheduling optimization according to the predicted value of the required computing power and the task allocation status of the edge node to obtain the edge node scheduling scheme;

[0080] Activate the target edge node according to the edge node scheduling scheme, and perform equipment state prediction based on the timing matrix of the lubrication system state deviation to obtain equipment state identification information;

[0081] When the equipment state identification information includes an abnormal state type, send the equipment state identification information to the management end through the cloud and perform synchronous display with the main collector.

[0082] It should be noted that in the above embodiments, the descriptions of the various embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0083] Those skilled in the art should understand that the embodiments of the present invention can be provided as a device, a system, or a computer program product. Therefore, the present invention can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0084] The present invention is described with reference to the flowcharts and / or block diagrams of the device, equipment (system), and computer program product according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram can be realized by computer program instructions, and the combination of the processes and / or blocks in the flowchart and / or block diagram can also be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for realizing the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0085] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured product including an instruction device, and the instruction device realizes the functions in the process Figure 1one or more processes and / or blocks Figure 1 the functions specified in one or more blocks.

[0086] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are performed on the computer or other programmable apparatus to produce a computer-implemented process, and thus the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one Figure 1 one or more processes and / or blocks Figure 1 or more processes and / or the functions specified in one or more blocks.

[0087] Although the preferred embodiments of the present invention have been described, additional changes and modifications can be made by those skilled in the art once they learn of the basic inventive concept.

[0088] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the present invention and its equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A device for the state monitoring and diagnosis of the lubrication system of active equipment, characterized in that, Deploy a system for monitoring and diagnosing the status of the lubrication system of a mobile device. The system includes a collection end, a cloud end, and a management end. The collection end includes a main collector and a slave collector. The execution steps include: After the mobile device is started and the timer meets the preset duration, the main collector receives the monitoring status of the lubrication system received by the slave collector, calculates the deviation from the theoretical status of the lubrication system to obtain the lubrication system status deviation time series matrix, and at the same time resets the timer to zero and starts timing again. Among them, the attributes of the lubrication system monitoring status include flow rate, viscosity, temperature, density, dielectric constant, water activity, water content, metal chip density, particle distribution density, oil-water interface level, oil-gas interface level, pressure, and degree of emulsification; Perform demand computing power prediction based on the lubrication system status deviation time series matrix to obtain the demand computing power prediction value; Obtain the task allocation status of the edge node; Perform edge node scheduling optimization according to the demand computing power prediction value and the edge node task allocation status to obtain the edge node scheduling scheme; Activate the target edge node according to the edge node scheduling scheme, and perform equipment status prediction based on the lubrication system status deviation time series matrix to obtain the equipment status identification information; When the equipment status identification information includes the abnormal status type, send the equipment status identification information to the management end through the cloud end and perform synchronous display with the main collector; Among them, performing demand computing power prediction based on the lubrication system status deviation time series matrix to obtain the demand computing power prediction value includes: Perform fuzzy deviation parameter analysis on the lubrication system status deviation time series matrix to obtain the first monitoring attribute fuzzy deviation parameter to the Nth monitoring attribute fuzzy deviation parameter; Extract the set of monitoring attributes with fuzzy deviation parameters greater than or equal to the fuzzy deviation threshold according to the first monitoring attribute fuzzy deviation parameter to the Nth monitoring attribute fuzzy deviation parameter; Perform demand computing power prediction according to the set of monitoring attributes to obtain the demand computing power prediction value; Among them, performing demand computing power prediction according to the set of monitoring attributes to obtain the demand computing power prediction value includes: Obtain the equipment status prediction unit library. Among them, the equipment status prediction unit library has several equipment status prediction units, and the several equipment status prediction units have several monitoring attribute weight distribution results and several demand computing power identifiers; Based on the set of monitoring attributes, traverse the several monitoring attribute weight distribution results for matching degree analysis to obtain several matching degrees; According to the several matching degrees, extract the set of target equipment status prediction units with matching degrees greater than or equal to the matching degree threshold from the equipment status prediction unit library; According to the set of target equipment status prediction units, perform demand computing power calculation based on the several demand computing power identifiers to obtain the demand computing power prediction value; Among them, based on the set of monitoring attributes, traversing the several monitoring attribute weight distribution results for matching degree analysis to obtain several matching degrees includes: Traverse the several monitoring attribute weight distribution results, sort the monitoring attributes from largest to smallest according to the weight to obtain several monitoring attribute sorting results; Obtain the set of monitored attributes, traverse the sequences of several monitored attributes, and extract the sequences of several monitored attribute serial numbers; Construct a matching degree evaluation function: where P(X i ) represents the matching degree of the i-th monitoring attribute serial number sequence, n represents the number of monitoring attribute sets, X ij represents the j-th serial number of the i-th monitoring attribute serial number sequence, and X i represents the i-th monitoring attribute serial number sequence; According to the matching degree evaluation function, traverse the sequences of several monitored attribute serial numbers for matching degree analysis to obtain the several matching degrees.

2. The device according to claim 1, characterized in that, Conduct fuzzy deviation parameter analysis based on the lubrication system state deviation time series matrix to obtain the first monitored attribute fuzzy deviation parameter to the Nth monitored attribute fuzzy deviation parameter, including: Obtain the set of monitored attribute deviation threshold values; Extract the first monitored attribute deviation threshold to the Nth monitored attribute deviation threshold of the set of monitored attribute deviation threshold values; Statistically analyze the proportion of moments when the deviation vector modulus in the first monitored attribute deviation vector modulus time series information of the lubrication system state deviation time series matrix is greater than or equal to the first monitored attribute deviation threshold, and set it as the first monitored attribute fuzzy deviation parameter; Until statistically analyze the proportion of moments when the deviation vector modulus in the Nth monitored attribute deviation vector modulus time series information of the lubrication system state deviation time series matrix is greater than or equal to the Nth monitored attribute deviation threshold, and set it as the Nth monitored attribute fuzzy deviation parameter.

3. The device according to claim 1, characterized in that, The steps for constructing the device state prediction unit library include: Match the active device operation logs according to the active device model, where the active device operation logs include the lubrication system state deviation time series record matrix and the active device state type record information; Obtain the preset weight distribution information of the monitored attributes; Configure the input layer of the long short-term memory neural network according to the preset weight distribution information of the monitored attributes. Using the active device state type record information as supervision and the lubrication system state deviation time series record matrix as the input, train the device state prediction unit and add it to the device state prediction unit library.

4. The device according to claim 1, characterized in that, Conduct edge node scheduling optimization according to the predicted demand computing power value and the edge node task allocation status to obtain an edge node scheduling plan, including: According to the edge node task allocation status, obtain the redundant computing power list of the edge node list; According to the predicted demand computing power value, based on the redundant computing power list, match the target edge node and add it to the edge node scheduling plan.

5. A system for state monitoring and diagnosis of the lubrication system of active equipment, characterized in that, Deployed in the device for monitoring and diagnosing the lubrication system state of active devices according to any one of claims 1 to 4.

Images