CN114298361A - Equipment state information prediction method and device - Google Patents
Equipment state information prediction method and device Download PDFInfo
- Publication number
- CN114298361A CN114298361A CN202011010322.6A CN202011010322A CN114298361A CN 114298361 A CN114298361 A CN 114298361A CN 202011010322 A CN202011010322 A CN 202011010322A CN 114298361 A CN114298361 A CN 114298361A
- Authority
- CN
- China
- Prior art keywords
- feature data
- data
- detection point
- target device
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method, a device, a readable medium and an electronic device for predicting device state information, wherein the method comprises the following steps: acquiring characteristic data of target equipment; generating a hash table of related data of the target device according to a hash function and the feature data of the target device, wherein the related data comprises feature data of detection points of non-target devices, and the feature data of the detection points are non-shared data; calculating the weight of the feature data of the detection point according to the hash table; establishing a joint model by using the weight of the characteristic data of the detection point and the equipment state label corresponding to the characteristic data of the detection point; and predicting the device state of the target device by using the combined model. According to the technical scheme provided by the invention, the data of the non-target equipment can be migrated to the target equipment, the relation between the characteristic data of the target equipment and the equipment state is established, the shared characteristic data between the equipment is not needed, and the data safety is ensured.
Description
Technical Field
The invention relates to the technical field of energy, in particular to a method and a device for predicting equipment state information.
Background
In order to monitor the industrial equipment to know the operation condition of the industrial equipment and further ensure the normal operation of the intelligent manufacturing system, equipment state prediction is generally required to be performed on the industrial equipment. Here, the device state prediction refers to predicting the probability of failure of the industrial device or the remaining service life of the industrial device from the characteristic data of the industrial device.
At present, historical operating data of a plurality of industrial devices are collected, a mapping relation between the historical operating data and the operating state of the industrial devices is established by adopting a machine learning method, an equipment state prediction model is obtained, and equipment state prediction of other industrial devices is realized by using the equipment state prediction model.
However, the above technical solutions need to share historical operating data of several industrial devices, thereby resulting in low data security among several industrial devices.
Disclosure of Invention
The present invention provides a method and an apparatus for predicting device status information, a computer-readable storage medium, and an electronic device, which are directed to the above technical problems in the prior art, and ensure data security without sharing feature data of devices between a non-target device and a target device.
In a first aspect, the present invention provides a method for predicting device status information, including:
acquiring characteristic data of target equipment;
generating a hash table of related data of the target device according to a hash function and the feature data of the target device, wherein the related data comprises feature data of detection points of non-target devices, and the feature data of the detection points are non-shared data;
calculating the weight of the feature data of the detection point according to the hash table;
establishing a joint model by using the weight of the characteristic data of the detection point and the equipment state label corresponding to the characteristic data of the detection point;
and predicting the device state of the target device by using the combined model.
In a second aspect, the present invention provides an apparatus for predicting device status information, including:
the data acquisition module is used for acquiring the characteristic data of the target equipment;
a generating module, configured to generate a hash table of relevant data of the target device according to a hash function and feature data of the target device, where the relevant data includes feature data of detection points of a non-target device, and the feature data of the detection points is non-shared data;
the weight calculation module is used for calculating the weight of the feature data of the detection point according to the hash table;
the model establishing module is used for establishing a combined model by using the weight of the characteristic data of the detection point and the equipment state label corresponding to the characteristic data of the detection point;
and the prediction module is used for predicting the device state of the target device by using the combined model.
In a third aspect, the invention provides a computer-readable storage medium comprising executable instructions which, when executed by a processor of an electronic device, perform the method according to any one of the first aspect or the method according to any one of the second aspect.
In a fourth aspect, the present invention provides an electronic device comprising a processor and a memory storing execution instructions, wherein when the processor executes the execution instructions stored in the memory, the processor performs the method according to any one of the first aspect or the method according to any one of the second aspect.
The invention provides a method, a device, a computer readable storage medium and an electronic device for predicting device state information, wherein the method comprises the steps of obtaining feature data of target devices, generating a hash table of related data of the target devices according to a hash function of the target devices and the feature data of the target devices, wherein the related data comprises the feature data of detection points of non-target devices, the feature data of the detection points are non-shared data, calculating the weight of the feature data of the detection points according to the hash table, establishing a combined model by using the weight of the feature data of the detection points and device state labels corresponding to the feature data of the detection points, and predicting the device state of the target devices by the combined model, and migrating the data to the target equipment, establishing the relation between the characteristic data of the target equipment and the equipment state, and ensuring the data safety without sharing the characteristic data between the equipment.
Further effects of the above-mentioned unconventional preferred modes will be described below in conjunction with specific embodiments.
Drawings
In order to more clearly illustrate the embodiments or the prior art solutions of the present invention, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments described in the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive labor.
Fig. 1 is a flowchart illustrating a method for predicting device status information according to an embodiment of the present invention;
fig. 2 is a flowchart illustrating another method for predicting device status information according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of an apparatus for predicting device status information according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be described in detail and completely with reference to the following embodiments and accompanying drawings. It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, an embodiment of the present invention provides a method for predicting device status information, including the following steps:
Specifically, the target equipment is any industrial equipment of the intelligent manufacturing system, which is not particularly limited in this embodiment of the present invention, and preferably, the equipment to be predicted is energy equipment, such as a gas boiler, an internal combustion engine, a steam turbine, a cogeneration equipment, a photovoltaic equipment, and the like.
Specifically, the target device has a plurality of feature data, each feature data includes a plurality of features respectively corresponding to feature values of the target device, where the plurality of features are influence factors influencing a device state, and it is specifically required to determine in combination with an actual scene, for example, if the target device is a gas boiler, the plurality of features include, but are not limited to, a gas flow, a gas temperature, a smoke exhaust temperature, a smoke flow, a gas pressure, an on-off state, a smoke humidity, and a smoke pressure.
Specifically, the step may be implemented by a detection point of the target device in the internet of things, where the detection point obtains a plurality of feature data of the target device through a sensor installed on the target device. The detection point of the target device may be understood as a node which is closest to the target device and is capable of performing data processing and data interaction, including but not limited to any one or more of an edge server, an edge gateway, and an edge controller.
102, generating a hash table of relevant data of the target device according to a hash function and the feature data of the target device, wherein the relevant data includes feature data of detection points of the non-target device, and the feature data of the detection points are non-shared data.
Specifically, for each feature data of the target device, converting the feature data through a hash function to obtain a hash value of the feature data; and obtaining a hash table through the hash values of all the characteristic data of the target equipment. The hash value may be understood as a value obtained by performing keyless encryption on feature data of the target device through a hash function. It should be noted that, in consideration of the fact that the hash function can rapidly and simply implement the keyless encryption processing of the data, the embodiment of the present invention selects the hash function to perform the keyless encryption on the feature data of the target device.
Specifically, the target device and the non-target device are the same device type, for example, the target device and the non-target device are both gas boilers, but information of production units, device models, and the like may be different.
Specifically, a detection point of the non-target device is a node of the non-target device in the internet of things, and the detection point acquires a plurality of feature data of the non-target device through a sensor mounted on the non-target device. The detection point may be understood as a node closest to the non-target device and capable of performing data processing and data interaction, including but not limited to any one or more of an edge server, an edge gateway, and an edge controller. It should be noted that the number of the non-target devices may be multiple, and each non-target device corresponds to one detection point.
Specifically, there are a plurality of feature data of the non-target device, each feature data includes a plurality of feature values corresponding to the non-target device, where the plurality of features corresponding to the feature data of the non-target device and the feature data of the target device are the same.
It should be noted that, if the feature data of the detection point is the non-shared data, it is indicated that the feature data of the detection point is not beyond the detection point, so that data security of the detection point is ensured, and meanwhile, data sharing is not required between the target device and the non-target device, so that data security is ensured.
And 103, calculating the weight of the feature data of the detection point according to the hash table.
In this embodiment, the data migration of the non-target device to the target device is established by detecting the weight of the feature data of the point, and the data association between the non-target device and the target device is established. Here, the feature data of the detection point is non-shared data, that is, the target device does not directly acquire the feature data of the detection point of the non-target device, that is, there is no data sharing between the target device and the non-target device, thereby ensuring data security.
In one embodiment, the weight of the feature data of the detection point may be determined specifically by:
determining a hash value of the feature data of the detection point; determining the distribution probability of the feature data of the detection points corresponding to the target equipment according to the hash table and the hash values of the feature data of the detection points; determining the distribution probability of the feature data of the detection points corresponding to the non-target equipment; and determining the ratio of the distribution probability of the feature data of the detection points corresponding to the target equipment to the distribution probability of the feature data of the detection points corresponding to the non-target equipment as the weight of the feature data of the detection points.
In the embodiment, the weight of the feature data of the detection point is determined according to the distribution probability of the feature data of the detection point corresponding to the target device and the distribution probability of the feature data of the detection point corresponding to the non-target device, so that the feature data of the detection point is migrated to the target device, and meanwhile, the feature data of the target device does not need to be shared, and the data safety is ensured.
Specifically, the hash value of the feature data of the detection point can be understood as a value obtained by encrypting the feature data of the detection point without a key, so that the feature data of the detection point does not need to be shared, only the hash value of the feature data of the detection point needs to be shared, and data security is ensured. It should be noted that, when the parameters in the hash function are different, the hash value of the feature data of the detection point and the analysis difficulty of the hash table may be increased, and therefore, the hash value of the feature data of the detection point is obtained by performing keyless encryption based on the hash function in step 101, thereby reducing the analysis difficulty between the hash value and the hash table.
Optionally, the hash value of the feature data of the detection point of the non-target device is obtained from the detection point, the detection point may be obtained by obtaining the feature data of the non-target device, and then, performing keyless encryption on each feature data based on the hash function in step 101 to obtain a hash value corresponding to each feature data of the detection point.
Optionally, the detection point of the non-target device sends the hash value of the feature data of the non-target device to the detection point of the target device, the detection point of the target device determines the distribution probability of the feature data of the non-target device corresponding to the target device according to the hash table and the hash value of the feature data of the non-target device, and sends the distribution probability to the detection point of the non-target device, the detection point of the non-target device determines, for each piece of feature data of the non-target device, the distribution probability of the feature data corresponding to the non-target device, and determines the weight of the feature data by using the ratio of the distribution probability of the feature data corresponding to the target device to the distribution probability corresponding to the non-target device.
Optionally, the replacement data of the feature data of the detection point is determined from the feature data of the target device according to the hash table and the hash value of the feature data of the detection point, and the distribution probability of the replacement data in the feature data of the target device is determined as the distribution probability of the feature data of the detection point corresponding to the target device. Here, the replacement data corresponding to the feature data of the detection point is determined from the plurality of feature data of the target device simply and quickly by the hash value and the hash table, so that the data association between the target device and the non-target device is established. As a possible implementation manner, based on a locality sensitive hashing algorithm, similarity analysis is performed on hash values of the hash table and feature data of the detection points to determine replacement data of the feature data of the detection points from the feature data of the target device, and the locality sensitive hashing algorithm can save calculation time and improve calculation efficiency. When a plurality of pieces of replacement data of the feature data of the detection point are determined from the feature data of the target device, one piece of replacement data may be randomly selected.
Optionally, the distribution probability is determined based on a non-parametric estimation method. Here, considering that the non-parameter estimation method does not need to rely on any prior assumption of data, but is determined based on the data itself, and is simpler than the parameter estimation method, the embodiment of the present invention selects the non-parameter estimation method to determine the distribution probability, and it should be further noted that the present invention does not intend to limit the non-parameter estimation method, and may specifically combine with the actual situation to determine, considering that the data amount of the feature data in the embodiment of the present invention is relatively large, in order to estimate the data distribution more accurately, the kernel density estimation method is preferred. In other words, based on the kernel density estimation method, the distribution probability is determined.
And 104, establishing a joint model by using the weight of the characteristic data of the detection point and the equipment state label corresponding to the characteristic data of the detection point.
In the embodiment, the combined model is established through the weights respectively corresponding to the characteristic data of the detection points and the equipment state labels corresponding to the characteristic data of the detection points, and the characteristic data of the target equipment does not need to be shared, so that the data safety is ensured.
It should be noted that the weight of the feature data of the detection point establishes a data relationship between the feature data of the target device and the feature data of the detection point of the non-target device, so that the feature data of the detection point of the non-target device is migrated to the target device, and data security is ensured.
It should also be noted that data of the target device and the non-target device are distributed at different detection points in the internet of things, and the data security problem is caused when model training is performed on shared data.
In one embodiment, the joint model may be specifically determined by:
determining a local model of the detection point by using the weight of the characteristic data of the detection point and the equipment state label of the characteristic data of the detection point; and establishing a combined model according to the local model of each detection point.
Specifically, model training is carried out based on the weight corresponding to the feature data of the detection points and the equipment state label, a local model is determined, and then fusion is carried out according to the local models of the multiple detection points to establish a combined model.
Specifically, the feature data of the detection point and the weight corresponding to the feature data of the detection point can realize the replacement of the feature data of the target device, and subsequently, the model parameter is adjusted through the weights corresponding to the plurality of feature data of the detection point, so that the adjusted model parameter can reflect the relationship between the feature data of the target device and the device state, the sharing of the feature data of the detection point is not involved, and the data security is ensured.
Specifically, the device status information tag may be fault information of the target device or remaining service life of the target device, for example, the device status information tag may be a fault or normal, may also be a fault type, and may also be a fault degree. It should be noted that, in the embodiments of the present invention, it is not intended to limit the method for obtaining the device status label corresponding to each feature data of the detection point, and the method may also be manual labeling, rule labeling, or cluster labeling, where the labeling methods are all in the prior art, and are not described in detail herein.
In one embodiment, model fusion is carried out after model iteration is carried out on the local models of all the detection points based on a joint learning algorithm, and a joint model is established.
Specifically, model training is carried out on detection points corresponding to the non-target equipment based on weights and equipment state labels respectively corresponding to characteristic data of the detection points to determine a local model, then, iteration is carried out on the local model of the target equipment based on joint learning algorithm on the detection points corresponding to the target equipment and the detection points corresponding to the non-target equipment, and a joint model is established according to the local model after the iteration is completed. The hash value of the feature data of the detection point is sent by the detection point corresponding to the non-target device, and meanwhile, the local model is iterated by adopting a joint learning algorithm, so that data sharing is not involved between the non-target device and the target device, and data safety is ensured.
Specifically, there are multiple non-target devices, each of which corresponds to one detection point, each of which is trained to obtain one local model, and the local models at the detection points have the same structure, but the model parameters may be different, for example, all are regression models. It should be noted that, in the standard machine learning method, feature data of a point is centrally detected on a machine or a data center. The joint learning enables the detection points corresponding to the non-target devices to cooperatively learn the prediction model of the detection point where the target device is located, meanwhile, the detection point where the non-target device is located can also store the characteristic data of the detection point, and in addition, the characteristic data of the detection point of the non-target device does not need to be shared.
In particular, model fusion may also be averaging of parameters in the model.
Specifically, the local model may be a neural network model or a regression model, and the specific need is determined by combining actual needs.
Specifically, the local model may be iterated as follows;
a1, performing model training according to the multiple feature data of the detection point, the equipment state label corresponding to each feature data and the weight corresponding to each feature data to determine a local model;
a2, judging whether the error of the local model meets the iteration condition, if so, determining the local model as a detection point of the final model sent to the target equipment, and if not, executing A3;
a3, sending the model parameters of the local model to the corresponding detection points of the target device;
and A4, receiving the updated model parameters sent by the detection points of the target device, adjusting the updated model parameters according to the plurality of characteristic data, the device state labels corresponding to the plurality of characteristic data and the weights corresponding to the plurality of characteristic data, so as to determine the adjusted model parameters, replacing the model parameters of the local model with the adjusted model parameters, and executing A2.
And 105, predicting the device state of the target device by using the combined model.
Specifically, the current characteristic data of the target device is collected and substituted into the combined model, and the device state of the target device can be determined.
In order to better understand the data processing procedure between the detection point of the target device and the detection point of the non-target device, for example, assume that the detection points corresponding to 3 non-target devices are set to A, B and C, respectively, the detection point of the target device is set to D, taking A as an example for explanation, A sends the hash value of the acquired feature data to D and receives the distribution probability of the feature data sent by D corresponding to the target device, determining the weight corresponding to the feature data by combining the distribution probability of the feature data corresponding to the non-target equipment, obtaining the weight corresponding to the feature data in B and C by B and C according to the similar processing procedure of A, then respectively carrying out model training by A, B, C, and D, carrying out model fusion on the local model sent from A, B, C to obtain a device state prediction model of the target device.
According to the technical scheme, the embodiment of the invention has at least the following effective effects: the data association between the non-target device and the target device is established through the weight of the feature data of the detection point of the non-target device, the data on the non-target device is migrated to the target device, meanwhile, a combined model is established based on the non-shared data of the detection point, the state prediction of the target device is achieved through the combined model, the feature data of the device does not need to be shared between the non-target device and the target device, and therefore data safety is guaranteed.
Fig. 1 shows only a basic embodiment of the method of the present invention, and based on this, certain optimization and expansion can be performed, and other preferred embodiments of the method can also be obtained.
To more clearly illustrate the technical solution of the present invention, please refer to fig. 2, an embodiment of the present invention provides another method for predicting device status information, and the present embodiment is further described with reference to specific application scenarios on the basis of the foregoing embodiment. In this embodiment, the method may specifically include the following steps:
The hash function is initialized. Assuming that the detection point corresponding to the target device in the internet of things is N, there are N non-target devices, the detection points corresponding to the internet of things are N1, N2, … … and nN respectively, N1, N2, … … and nN respectively can obtain the feature data of the corresponding non-target devices, N can obtain all the feature data of the target device, and N performs keyless encryption on each feature data of the target device according to a hash function to obtain a hash table and sends the hash function to N1, N2, … … and nN. It should be noted that the process of obtaining the local model by n1, n2, … … and nN is similar, and the following description is given by the processing procedure of n 1.
n1 receives the hash function sent by n, encrypts the feature data of the non-target device without a key according to the hash function to obtain the hash value of the feature data of the detection point, and sends the hash value of the feature data of the detection point to n.
And n, receiving the hash value of the feature data of the detection point, and performing similarity analysis on the hash table and the hash value of the feature data of the detection point based on a locality sensitive hash algorithm to determine the replacement data of the feature data of the detection point from each feature data of the target device, so as to establish data relation between the target device and the non-target device.
n, based on the kernel density estimation algorithm, determining the distribution probability of each feature data of the replacement data in the target device, determining the distribution probability as the distribution probability that the feature data of the detection point corresponds to the target device, and sending the distribution probability that the feature data of the detection point corresponds to the target device to n 1.
n1 receives the distribution probability corresponding to the feature data of the detection point sent by n, determines the distribution probability of the feature data of the detection point corresponding to the non-target device based on a kernel density estimation method, and determines the ratio of the distribution probability of the feature data of the detection point corresponding to the target device and the distribution probability corresponding to the non-target device as the weight of the feature data of the detection point.
Here, the feature data of the detected points corresponds to the distribution probability in the non-target device, and it can be understood that the feature data of the detected points corresponds to the distribution probability in all the feature data of the non-target device at the detected points.
n1, performing model training according to the weight of the feature data of the detection point and the device state label, and determining a local model of the target device. n2, … … and nN respectively determine the local model of the target equipment according to a method similar to n 1.
And step 206, performing model fusion after model iteration on the local models of the detection points based on a joint learning algorithm, and establishing a joint model.
And n, n1, n2, … … and nN adopt a joint learning algorithm to carry out model iteration on local models obtained by respectively training n1, n2, n … … and nN, and then the local models are sent to n, and n fuses the local models which are obtained by respectively training n1, n2, n … … and nN and are subjected to iteration, so that a joint model of the target equipment is established.
And step 207, predicting the device state of the target device by using the combined model.
And n, predicting the equipment state of the target equipment according to the combined model.
According to the technical scheme, the beneficial effects of the embodiment are as follows: the establishment of a combined model of the target equipment is realized through detection points of the target equipment and the non-target equipment in the Internet of things, meanwhile, the non-target equipment replaces characteristic data with a hash value to share information with the target equipment, the sharing of the characteristic data is not involved between the detection points of the target equipment and the non-target equipment, and the data safety is ensured.
Based on the same concept as the method for predicting device status information provided in the embodiment of the method of the present invention, please refer to fig. 3, the present invention implements a device for predicting device status information, which is applied to a joint detection point of a target device, and includes:
a data obtaining module 301, configured to obtain feature data of a target device;
a generating module 302, configured to generate a hash table of relevant data of the target device according to a hash function and feature data of the target device, where the relevant data includes feature data of a detection point of a non-target device, and the feature data of the detection point is non-shared data;
a weight calculation module 303, configured to calculate a weight of the feature data of the detection point according to the hash table;
a model establishing module 304, configured to establish a joint model by using the weight of the feature data of the detection point and the device state label corresponding to the feature data of the detection point;
a prediction module 305, configured to perform device state prediction on the target device using the joint model.
In an embodiment of the present invention, the weight determining module 303 includes: a hash value determination unit, a first probability determination unit, a second probability determination unit, and a weight determination unit; wherein the content of the first and second substances,
the hash value determining unit is used for determining the hash value of the feature data of the detection point;
the first probability determination unit is configured to determine, according to the hash table and hash values of the feature data of the detection points, a distribution probability that the feature data of the detection points correspond to the target device;
the second probability determining unit is used for determining the distribution probability of the feature data of the detection points corresponding to the non-target equipment;
the weight determination unit is configured to determine a ratio of a distribution probability of the feature data of the detection point corresponding to the target device to a distribution probability of the feature data of the detection point corresponding to the non-target device as a weight of the feature data of the detection point.
In one embodiment of the present invention, the first probability determination unit includes: selecting a subunit and a probability determining subunit; wherein the content of the first and second substances,
the selecting subunit is configured to determine, according to the hash table and the hash value of the feature data of the detection point, replacement data of the feature data of the detection point from the feature data of the target device;
the probability determining subunit is configured to determine, as the distribution probability of the feature data of the detection point corresponding to the target device, the distribution probability of the replacement data in the feature data of the target device.
In an embodiment of the present invention, the selecting subunit is configured to perform, based on a locality sensitive hashing algorithm, a similarity analysis on the hash table and the hash value of the feature data of the detection point, so as to determine, from the feature data of the target device, replacement data of the feature data of the detection point.
In an embodiment of the invention, the hash value of the feature data of the detection point is determined according to the hash function.
In an embodiment of the present invention, the model building module 304 includes: a first model determining unit and a second model determining unit;
the first model determining unit is configured to determine a local model of the detection point by using the weight of the feature data of the detection point and the device state label of the feature data of the detection point;
and the second model determining unit is used for establishing a combined model according to the local model of each detection point.
In an embodiment of the present invention, the second model determining unit is configured to perform model fusion after performing model iteration on the local models of the detection points based on a joint learning algorithm, and establish a joint model.
Fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present invention. On the hardware level, the electronic device includes a processor 401 and a memory 402 storing execution instructions, and optionally an internal bus 403 and a network interface 404. The Memory 402 may include a Memory 4021, such as a Random-Access Memory (RAM), and may further include a non-volatile Memory 4022 (e.g., at least 1 disk Memory); the processor 401, the network interface 404, and the memory 402 may be connected to each other by an internal bus 403, and the internal bus 403 may be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, an EISA (Extended Industry Standard Architecture) bus, or the like; the internal bus 403 may be divided into an address bus, a data bus, a control bus, etc., which is indicated by only one double-headed arrow in fig. 4 for convenience of illustration, but does not indicate only one bus or one type of bus. Of course, the electronic device may also include hardware required for other services. When the processor 401 executes execution instructions stored by the memory 402, the processor 401 performs the method in any of the embodiments of the present invention and at least is used to perform the method as shown in fig. 1 or fig. 2.
In a possible implementation manner, the processor reads the corresponding execution instruction from the nonvolatile memory to the memory and then runs the corresponding execution instruction, and the corresponding execution instruction can also be obtained from other equipment, so as to form a prediction device of the equipment state information on a logic level. The processor executes the execution instructions stored in the memory, so that the prediction method of the device state information provided by any embodiment of the invention is realized through the executed execution instructions.
The processor may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware in a processor or instructions in the form of software. The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but also Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components. The various methods, steps and logic blocks disclosed in the embodiments of the present invention may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
Embodiments of the present invention further provide a computer-readable storage medium, which includes an execution instruction, and when a processor of an electronic device executes the execution instruction, the processor executes a method provided in any one of the embodiments of the present invention. The electronic device may specifically be the electronic device shown in fig. 4; the execution instruction is a computer program corresponding to the prediction method of the equipment state information.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects.
The multiple embodiments of the present invention are described in a progressive manner, and the same and similar parts among the multiple embodiments can be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, as for the apparatus embodiment, since it is substantially similar to the method embodiment, the description is relatively simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The above description is only an example of the present invention, and is not intended to limit the present invention. Various modifications and alterations to this invention will become apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.
Claims (10)
1. A method for predicting device state information, comprising:
acquiring characteristic data of target equipment;
generating a hash table of related data of the target device according to a hash function and the feature data of the target device, wherein the related data comprises feature data of detection points of non-target devices, and the feature data of the detection points are non-shared data;
calculating the weight of the feature data of the detection point according to the hash table;
establishing a joint model by using the weight of the characteristic data of the detection point and the equipment state label corresponding to the characteristic data of the detection point;
and predicting the device state of the target device by using the combined model.
2. The method of claim 1, wherein the calculating the weight of the feature data of the detection point according to the hash table comprises:
determining a hash value of the feature data of the detection point;
determining the distribution probability of the feature data of the detection points corresponding to the target equipment according to the hash table and the hash value of the feature data of the detection points;
determining the distribution probability of the feature data of the detection points corresponding to the non-target equipment;
and determining the ratio of the distribution probability of the feature data of the detection points corresponding to the target equipment to the distribution probability of the feature data of the detection points corresponding to the non-target equipment as the weight of the feature data of the detection points.
3. The method of claim 2, wherein the determining the distribution probability that the feature data of the detection points correspond to the target device according to the hash table and the hash value of the feature data of the detection points comprises:
determining replacement data of the feature data of the detection point from the feature data of the target device according to the hash table and the hash value of the feature data of the detection point;
and determining the distribution probability of the replacement data in the feature data of the target device as the distribution probability of the feature data of the detection point corresponding to the target device.
4. The method according to claim 3, wherein the determining replacement data for the feature data of the detection point from the feature data of the target device based on the hash table and the hash value of the feature data of the detection point comprises:
and based on a locality sensitive hashing algorithm, performing similarity analysis on the hash table and the hash value of the feature data of the detection point to determine replacement data of the feature data of the detection point from the feature data of the target device.
5. The method according to claim 2, characterized in that the hash value of the feature data of the detection points is determined according to the hash function.
6. The method of claim 1, wherein the building a joint model by using the weight of the feature data of the detection point and the device state label corresponding to the feature data of the detection point comprises:
determining a local model of the detection point by using the weight of the characteristic data of the detection point and the equipment state label of the characteristic data of the detection point;
and establishing a combined model according to the local model of each detection point.
7. The method of claim 6, wherein said building a joint model from the local models of the respective detection points comprises:
and performing model fusion after performing model iteration on the local models of the detection points based on a joint learning algorithm, and establishing a joint model.
8. An apparatus for predicting device status information, comprising:
the data acquisition module is used for acquiring the characteristic data of the target equipment;
a generating module, configured to generate a hash table of relevant data of the target device according to a hash function and feature data of the target device, where the relevant data includes feature data of detection points of a non-target device, and the feature data of the detection points is non-shared data;
the weight calculation module is used for calculating the weight of the feature data of the detection point according to the hash table;
the model establishing module is used for establishing a combined model by using the weight of the characteristic data of the detection point and the equipment state label corresponding to the characteristic data of the detection point;
and the prediction module is used for predicting the device state of the target device by using the combined model.
9. A readable medium comprising executable instructions which, when executed by a processor of an electronic device, cause the electronic device to perform the method of any of claims 1 to 7.
10. An electronic device comprising a processor and a memory storing execution instructions, the processor performing the method of any of claims 1-7 when the processor executes the execution instructions stored by the memory.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011010322.6A CN114298361A (en) | 2020-09-23 | 2020-09-23 | Equipment state information prediction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011010322.6A CN114298361A (en) | 2020-09-23 | 2020-09-23 | Equipment state information prediction method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114298361A true CN114298361A (en) | 2022-04-08 |
Family
ID=80964124
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011010322.6A Pending CN114298361A (en) | 2020-09-23 | 2020-09-23 | Equipment state information prediction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114298361A (en) |
-
2020
- 2020-09-23 CN CN202011010322.6A patent/CN114298361A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111414999B (en) | Method and device for monitoring running state of equipment | |
| CN110875851B (en) | Underground cable fault early warning method and device | |
| CN108989468B (en) | Trust network construction method and device | |
| CN112330060A (en) | Equipment fault prediction method and device, readable storage medium and electronic equipment | |
| KR102438527B1 (en) | Method for predicting malignant url considering url- related content and host data and apparatus implementing the same method | |
| CN111143165A (en) | Monitoring method and device | |
| CN114595826A (en) | Method, system, terminal and storage medium for selecting nodes of federated learner | |
| CN110930254A (en) | Data processing method, device, terminal and medium based on block chain | |
| CN110515631B (en) | Method for generating application installation data packet, server and computer storage medium | |
| CN111461925A (en) | Power distribution network fault analysis method, device, equipment and storage medium | |
| CN110826606A (en) | Element matching method, device, server and readable storage medium | |
| CN114298361A (en) | Equipment state information prediction method and device | |
| CN112330063A (en) | Equipment failure prediction method, device and computer readable storage medium | |
| CN114429235A (en) | Equipment fault prediction method and device, readable medium and electronic equipment | |
| CN114648050A (en) | Equipment fault prediction method and device, readable storage medium and electronic equipment | |
| CN105227589B (en) | File downloading method and device | |
| CN112866234B (en) | Network attack tracing method, device and system | |
| CN112733453B (en) | Equipment predictive maintenance method and device based on joint learning | |
| CN115622264A (en) | Power information monitoring system and method | |
| CN114254791A (en) | Method and device for predicting oxygen content of flue gas | |
| CN113556353A (en) | Big data based information monitoring and reminding method and artificial intelligence cloud service system | |
| CN111030091B (en) | Method and system for determining installed electric capacity of distributed renewable energy | |
| CN114254792A (en) | Method and device for predicting oxygen content of flue gas | |
| WO2015135182A1 (en) | Method and apparatus for evaluating performance of real-time system | |
| CN114429795A (en) | Method and device for predicting oxygen content of boiler flue gas |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |