CN110658905A - Early warning method, early warning system and early warning device for equipment running state - Google Patents
Early warning method, early warning system and early warning device for equipment running state Download PDFInfo
- Publication number
- CN110658905A CN110658905A CN201910898842.6A CN201910898842A CN110658905A CN 110658905 A CN110658905 A CN 110658905A CN 201910898842 A CN201910898842 A CN 201910898842A CN 110658905 A CN110658905 A CN 110658905A
- Authority
- CN
- China
- Prior art keywords
- state data
- data
- historical
- component
- current
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
- G06F1/26—Power supply means, e.g. regulation thereof
- G06F1/28—Supervision thereof, e.g. detecting power-supply failure by out of limits supervision
-
- 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
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The disclosure provides an early warning method, an early warning system, an early warning device and a computer readable storage medium of an equipment running state, and relates to the technical field of information. The early warning method for the running state of the equipment comprises the following steps: acquiring historical state data and current state data of equipment; obtaining a predicted value of the current state data according to the historical state data; judging the running state of the equipment according to the current state data and the predicted value; and sending an early warning signal to the equipment according to the running state of the equipment. The method and the device can judge the running state of the equipment more accurately and realize accurate early warning of the running state of the equipment.
Description
Technical Field
The present disclosure relates to the field of information technologies, and in particular, to an early warning method, an early warning system, an early warning apparatus, and a computer-readable storage medium for an operating state of a device.
Background
In recent years, the concept of internet + has also been proposed as the internet age has come. Colloquially, the internet plus is the internet plus all traditional industries. However, the above-mentioned expressions do not refer to simple addition, but use information technology and internet platform to make the internet and traditional industry deeply merge to create new development ecology.
Currently, with the frequent emergence of various large energy enterprises in the world, the development ecology of energy interconnection is gradually formed. Energy interconnection and energy management, and early warning of the running state of the equipment can not be avoided.
Disclosure of Invention
The technical problem solved by the present disclosure is how to realize accurate early warning of the running state of the device.
According to an aspect of the embodiments of the present disclosure, there is provided a method for early warning of an operating state of a device, including: acquiring historical state data and current state data of equipment; obtaining a predicted value of the current state data according to the historical state data; judging the running state of the equipment according to the current state data and the predicted value; and sending an early warning signal to the equipment according to the running state of the equipment.
In some embodiments, obtaining the predicted value of the current state data from the historical state data comprises: performing local mean decomposition on the historical state data in the time series form to obtain a plurality of PF components and a residual component of the historical state data; respectively inputting each PF component and residual component of the historical state data into a corresponding pre-trained time convolution neural network to obtain the predicted value of each PF component and residual component of the current state data; and determining the predicted value of the current state data according to the predicted values of each PF component and residual component of the current state data.
In some embodiments, determining the predicted value of the current state data according to the predicted values of the PF components and the residual components of the current state data includes: respectively determining cross-correlation coefficients between each PF component and residual component of the historical state data and the historical state data in a time series form; and determining the predicted value of the current state data according to the cross correlation coefficient, the predicted values of each PF component and the residual component of the current state data.
In some embodiments, determining the predicted value of the current state data from the cross-correlation coefficient, the respective PF components of the current state data, and the predicted value of the residual component comprises: determining a prediction value for current state data in the following manner
Wherein c represents the predicted value of the current state data, n represents the total number of PF components, i represents the identity of PF components, aiRepresenting the cross-correlation coefficient between the ith PF component and the historical state data in the form of a time series, the PFiDenotes the i-th PF component, b denotes the cross-correlation coefficient between the residual component and the historical state data in the form of a time series, and S denotes the residual component.
In some embodiments, obtaining the predicted value of the current state data from the historical state data further comprises: performing local mean decomposition on training data in a time series form to obtain a plurality of PF components and a residual component of the training data; and respectively training corresponding time convolution neural networks by using each PF component and residual component of the training data, so that each trained time convolution neural network can output the predicted value of the PF component according to the input PF component or output the predicted value of the residual component according to the input residual component.
In some embodiments, obtaining historical state data and current state data of the device comprises: periodically acquiring historical state data and current state data of equipment; obtaining the predicted value of the current state data according to the historical state data comprises: periodically obtaining a predicted value of the current state data according to the historical state data; according to the current state data and the predicted value, judging the running state of the equipment comprises the following steps: periodically determining the deviation between the current state data and the predicted value; and under the condition that the times of the deviation larger than the first threshold is larger than the second threshold, judging that the running state of the equipment is abnormal.
In some embodiments, sending the alert signal to the device based on the operational state of the device comprises: and sending a power failure early warning signal to the equipment under the condition that the running state of the equipment is abnormal.
In some embodiments, determining the operating state of the device based on the current state data and the predicted value comprises: judging a first running state of the equipment according to the current state data and the predicted value; clustering the historical state data and the current state data, and labeling an operating state label for each data cluster obtained by clustering; determining a second running state of the equipment according to the running state label of the data cluster to which the current state data belongs; according to the running state of the equipment, sending an early warning signal to the equipment comprises the following steps: and sending a power failure early warning signal to the equipment under the condition that at least one of the first running state and the second running state is abnormal.
In some embodiments, the historical state data and the current state data comprise at least one of: historical operating voltage and current operating voltage; historical operating voltage and current operating current; historical operating power and current operating power; historical power consumption and current power consumption.
According to another aspect of the embodiments of the present disclosure, there is provided an early warning system for an operating state of a device, including: the data transmission module is configured to acquire historical state data and current state data of the equipment; the data prediction module is configured to obtain a prediction value of the current state data according to the historical state data; the state judgment module is configured to judge the running state of the equipment according to the current state data and the predicted value; and the early warning decision module is configured to send an early warning signal to the equipment according to the running state of the equipment.
In some embodiments, the data prediction module is configured to: performing local mean decomposition on the historical state data in the time series form to obtain a plurality of PF components and a residual component of the historical state data; respectively inputting each PF component and residual component of the historical state data into a corresponding pre-trained time convolution neural network to obtain the predicted value of each PF component and residual component of the current state data; and determining the predicted value of the current state data according to the predicted values of each PF component and residual component of the current state data.
In some embodiments, the data prediction module is configured to: respectively determining cross-correlation coefficients between each PF component and residual component of the historical state data and the historical state data in a time series form; and determining the predicted value of the current state data according to the cross correlation coefficient, the predicted values of each PF component and the residual component of the current state data.
In some embodiments, the data prediction module is configured to: determining a prediction value for current state data in the following manner
Wherein c represents the predicted value of the current state data, n represents the total number of PF components, i represents the identity of PF components, aiRepresenting the cross-correlation coefficient between the ith PF component and the historical state data in the form of a time series, the PFiDenotes the i-th PF component, b denotes the cross-correlation coefficient between the residual component and the historical state data in the form of a time series, and S denotes the residual component.
In some embodiments, the data prediction module is further configured to: performing local mean decomposition on training data in a time series form to obtain a plurality of PF components and a residual component of the training data; and respectively training corresponding time convolution neural networks by using each PF component and residual component of the training data, so that each trained time convolution neural network can output the predicted value of the PF component according to the input PF component or output the predicted value of the residual component according to the input residual component.
In some embodiments, the data transmission module is configured to: periodically acquiring historical state data and current state data of equipment; the data prediction module is configured to: periodically obtaining a predicted value of the current state data according to the historical state data; the state determination module is configured to: periodically determining the deviation between the current state data and the predicted value; and under the condition that the times of the deviation larger than the first threshold is larger than the second threshold, judging that the running state of the equipment is abnormal.
In some embodiments, the early warning decision module is configured to: and sending a power failure early warning signal to the equipment under the condition that the running state of the equipment is abnormal.
In some embodiments, the state determination module is configured to: judging a first running state of the equipment according to the current state data and the predicted value; clustering the historical state data and the current state data, and labeling an operating state label for each data cluster obtained by clustering; determining a second running state of the equipment according to the running state label of the data cluster to which the current state data belongs; the early warning decision module is configured to: and sending a power failure early warning signal to the equipment under the condition that at least one of the first running state and the second running state is abnormal.
In some embodiments, the historical state data and the current state data comprise at least one of: historical operating voltage and current operating voltage; historical operating voltage and current operating current; historical operating power and current operating power; historical power consumption and current power consumption.
According to another aspect of the embodiments of the present disclosure, there is provided an early warning apparatus for an operating state of a device, including: a memory; and a processor coupled to the memory, the processor configured to execute the aforementioned method for early warning of the device operating state based on instructions stored in the memory.
According to still another aspect of the embodiments of the present disclosure, a computer-readable storage medium is provided, where the computer-readable storage medium stores computer instructions, and the instructions, when executed by a processor, implement the foregoing method for early warning of the device operation state.
According to another aspect of the embodiments of the present disclosure, there is provided an early warning apparatus for an operating state of a device, including: a memory; and a processor coupled to the memory, the processor configured to execute the aforementioned method for early warning of the device operating state based on instructions stored in the memory.
According to still another aspect of the embodiments of the present disclosure, a computer-readable storage medium is provided, where the computer-readable storage medium stores computer instructions, and the instructions, when executed by a processor, implement the foregoing method for early warning of the device operation state.
The method and the device can judge the running state of the equipment more accurately and realize accurate early warning of the running state of the equipment.
Other features of the present disclosure and advantages thereof will become apparent from the following detailed description of exemplary embodiments thereof, which proceeds with reference to the accompanying drawings.
Drawings
In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only some embodiments of the present disclosure, and for those skilled in the art, other drawings can be obtained according to the drawings without inventive exercise.
Fig. 1 shows a flow chart of a method for warning an operation state of a device according to some embodiments of the present disclosure.
FIG. 2 illustrates a flow diagram of some embodiments for obtaining a prediction value for current state data from historical state data.
Fig. 3 shows a schematic structural diagram of an early warning system for the device operation state according to some embodiments of the present disclosure.
Fig. 4 shows a schematic structural diagram of a specific application example of the early warning system for the operation state of the device of the present disclosure.
Fig. 5 is a schematic structural diagram of an early warning device for an operating state of an apparatus according to some embodiments of the present disclosure.
Detailed Description
The technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the drawings in the embodiments of the present disclosure, and it is obvious that the described embodiments are only a part of the embodiments of the present disclosure, and not all of the embodiments. The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the disclosure, its application, or uses. All other embodiments, which can be derived by a person skilled in the art from the embodiments disclosed herein without making any creative effort, shall fall within the protection scope of the present disclosure.
Some embodiments of the method for warning the operation state of the device of the present disclosure are described first with reference to fig. 1.
Fig. 1 shows a flow chart of a method for warning an operation state of a device according to some embodiments of the present disclosure. As shown in fig. 1, the present embodiment includes steps S101 to S104.
In step S101, history status data and current status data of the device are acquired.
Those skilled in the art will understand that the above-mentioned device may be specifically an energy device applied in many fields such as wind power. In the case of an energy device, during the process of acquiring status data, historical status data and current status data of the device may be acquired periodically, and these status data may include voltage, current, power consumption, and so on. That is, the historical state data and the current state data include at least one of: historical operating voltage and current operating voltage, historical operating voltage and current operating current, historical operating power and current operating power, historical power consumption and current power consumption.
In step S102, a prediction value of the current state data is obtained from the historical state data.
In the process of obtaining the predicted value of the current state data, the predicted value of the current state data can be periodically obtained according to the historical state data. For example, the current state data at time t +1 may be predicted from the historical state data at time t before the energy device. And then, after waiting for obtaining the actual value of the state data of the energy equipment at the t +1 moment, predicting the current state data of the energy equipment at the t +2 moment according to the historical state data of the energy equipment at the previous t +1 moment, and so on.
In step S103, the operating state of the equipment is determined based on the current state data and the predicted value.
In the process of judging the running state of the equipment, the deviation between the current state data and the predicted value can be periodically determined, and the running state of the equipment is judged to be abnormal under the condition that the frequency of the deviation being greater than the first threshold is greater than the second threshold. For example, after acquiring the actual value of the state data at the time t +1 of the energy device, the actual value at the time t +1 is compared with the predicted value at the time t + 1. And after the actual value of the state data of the energy equipment at the time t +2 is obtained, comparing the actual value of the state data at the time t +2 with the predicted value at the time t +2, and so on. And if the deviation between the actual value and the predicted value is large for five times continuously, judging that the running state of the equipment is abnormal.
In step S104, an early warning signal is sent to the device according to the operating state of the device.
For example, in the case that the device is in a normal operation state, the warning signal may not be sent to the device; under the condition that the running state of the equipment is abnormal, a power failure early warning signal can be sent to the equipment to inform that the equipment has a fault and needs power failure renovation.
According to the method and the device, the running state of the device can be judged more accurately by using the historical state data and the current state data of the device, so that more accurate early warning of the running state of the device is realized.
Some embodiments for obtaining a prediction value for current state data from historical state data are described below in conjunction with FIG. 2.
FIG. 2 illustrates a flow diagram of some embodiments for obtaining a prediction value for current state data from historical state data. As shown in fig. 2, the present embodiment includes steps S2021 to S2024.
In step S2021, local mean decomposition is performed on the training data in the time series form to obtain a plurality of PF components and a residual component of the training data.
The time series form training data is decomposed by an LMD (Local Mean Decomposition) method, and a plurality of PF components and a residual component can be obtained. Each PF component may implicitly represent seasonal factors, meteorological factors, artifacts, stochastic factors, etc., and the residual component may implicitly represent the trend of the change in the device status data. Through LMD decomposition, the influence trend of the factors on the equipment state data is decomposed and considered.
When there are a plurality of types of data in the history state data, the LMD decomposition may be performed for each type of data (voltage data U, current data I, power data P, power consumption data Q), respectively. For example, LMD decomposition of the voltage training data in time series may result in PF components and a residual component of the voltage training data.
In step S2022, each PF component and residual component of the training data is used to train a corresponding time convolution neural network, so that each time convolution neural network after training can output a predicted value of the PF component according to the input PF component or output a predicted value of the residual component according to the input residual component.
The relevant content of the Time-convolutional neural network Time-CNN can be referred to as a review for Time series classification (Data Mining and Knowledge Discovery). And the Time-CNN can be constructed in an auxiliary manner from the qualitative aspect by utilizing each PF component and each residual component. Because the data input by the time convolution neural network is in the form of time series, the training can be carried out in an unsupervised learning mode. Still taking the voltage data as an example, the Time convolution neural network 1 is obtained by adopting PF component 1 training of the voltage training data, the Time convolution neural network 2 … … is obtained by adopting PF component n training of the voltage training data, the Time convolution neural network n is obtained by adopting residual component training of the voltage training data, and then the trained Time-CNN models are stored for use in prediction. The non-linear fitting capability of the Time-CNN can save the influence capability of influencing factors of historical data and qualitative expression on future data change in the Time-CNN model.
In step S2023, local mean decomposition is performed on the historical state data in the form of time series to obtain a plurality of PF components and a residual component of the historical state data.
When there are a plurality of types of data in the history state data, the LMD decomposition may be performed for each type of data (voltage data U, current data I, power data P, power consumption data Q), respectively. For example, LMD decomposition of the voltage data U in time series may result in a plurality of PF components and a residual component of the voltage history status data.
In step S2024, the PF components and the residual components of the historical state data are input into the corresponding pre-trained time convolution neural networks, respectively, so as to obtain the predicted values of the PF components and the residual components of the current state data.
For example, the PF component 1 of the voltage history state data is input to the time convolution neural network 1, the PF component 2 of the voltage history state data is input to the time convolution neural network 2 … …, the PF component n of the voltage history state data is input to the time convolution neural network n, and the residual component of the voltage history state data is input to the time convolution neural network n +1, so that the predicted values of each PF component and residual component of the current state data can be obtained.
In step S2025, the predicted value of the current state data is determined based on the predicted values of the PF components and the residual components of the current state data.
In the process of determining the predicted value of the current state data, cross correlation coefficients between each PF component and residual component of the historical state data and the historical state data in a time series form are determined respectively. And then determining the predicted value of the current state data according to the cross correlation coefficient, the predicted values of each PF component and the residual component of the current state data.
For example, the predicted value of the current state data may be determined in the following manner
Wherein c represents the predicted value of the current state data, n represents the total number of PF components, i represents the identity of PF components, aiRepresenting the cross-correlation coefficient between the ith PF component and the historical state data in the form of a time series, the PFiDenotes the i-th PF component, b denotes the cross-correlation coefficient between the residual component and the historical state data in the form of a time series, and S denotes the residual component.
According to the method and the device, the LMD decomposition and the Time-CNN are jointly adopted to predict the current operation state data of the equipment, and the influence between different factors implied in the Time sequence of historical state data is reduced, so that the accuracy of predicting the current operation state data of the equipment is improved, the more accurate judgment of the operation state of the equipment is facilitated, the more accurate early warning of the operation state of the equipment is realized, and the robustness of the early warning are improved.
The corresponding embodiment in fig. 2 describes a method for implementing step S102 in the corresponding embodiment in fig. 1. An alternative implementation of the further steps in the embodiment corresponding to fig. 1 is described below.
Step S103 in the embodiment corresponding to fig. 1 may include: judging a first running state of the equipment according to the current state data and the predicted value; clustering the historical state data and the current state data, and labeling an operating state label for each data cluster obtained by clustering; and determining a second running state of the equipment according to the running state label of the data cluster to which the current state data belongs. Step S104 in the corresponding embodiment of fig. 1 may include: and sending a power failure early warning signal to the equipment under the condition that at least one of the first running state and the second running state is abnormal.
The voltage state data is still taken as an example. And supposing that the first running state of the equipment is judged to be normal voltage according to the current voltage state data and the predicted value. Then, clustering the historical voltage state data and the current voltage state data, and labeling an operating state label for each data cluster obtained by clustering, wherein the label can comprise normal voltage and abnormal voltage. And determining the second operation state of the equipment as voltage abnormity according to the operation state label of the data cluster to which the current state data of the voltage belongs. And sending a power failure early warning signal to the equipment according to the first operation state as normal voltage and the second operation state as abnormal voltage.
Some embodiments of the warning system for the operation status of the device of the present disclosure are described below with reference to fig. 3.
Fig. 3 shows a schematic structural diagram of an early warning system for the device operation state according to some embodiments of the present disclosure. As shown in fig. 3, the early warning system 30 for the device operation state in the present embodiment includes: a data transmission module 301 configured to acquire historical status data and current status data of a device; a data prediction module 302 configured to obtain a prediction value of the current state data according to the historical state data; a state judgment module 303 configured to judge an operation state of the device according to the current state data and the predicted value; and the early warning decision module 304 is configured to send an early warning signal to the equipment according to the running state of the equipment.
In some embodiments, the data prediction module 302 is configured to: performing local mean decomposition on the historical state data in the time series form to obtain a plurality of PF components and a residual component of the historical state data; respectively inputting each PF component and residual component of the historical state data into a corresponding pre-trained time convolution neural network to obtain the predicted value of each PF component and residual component of the current state data; and determining the predicted value of the current state data according to the predicted values of each PF component and residual component of the current state data.
In some embodiments, the data prediction module 302 is configured to: respectively determining cross-correlation coefficients between each PF component and residual component of the historical state data and the historical state data in a time series form; and determining the predicted value of the current state data according to the cross correlation coefficient, the predicted values of each PF component and the residual component of the current state data.
In some embodiments, the data prediction module 302 is configured to: determining a prediction value for current state data in the following manner
Wherein c represents the predicted value of the current state data, n represents the total number of PF components, i represents the identity of PF components, aiRepresenting the cross-correlation coefficient between the ith PF component and the historical state data in the form of a time series, the PFiDenotes the i-th PF component, b denotes the cross-correlation coefficient between the residual component and the historical state data in the form of time seriesAnd S denotes a residual component.
In some embodiments, the data prediction module 302 is further configured to: performing local mean decomposition on training data in a time series form to obtain a plurality of PF components and a residual component of the training data; and respectively training corresponding time convolution neural networks by using each PF component and residual component of the training data, so that each trained time convolution neural network can output the predicted value of the PF component according to the input PF component or output the predicted value of the residual component according to the input residual component.
In some embodiments, the data transmission module 301 is configured to: periodically acquiring historical state data and current state data of equipment; the data prediction module 302 is configured to: periodically obtaining a predicted value of the current state data according to the historical state data; the state determination module 303 is configured to: periodically determining the deviation between the current state data and the predicted value; and under the condition that the times of the deviation larger than the first threshold is larger than the second threshold, judging that the running state of the equipment is abnormal.
In some embodiments, the early warning decision module 304 is configured to: and sending a power failure early warning signal to the equipment under the condition that the running state of the equipment is abnormal.
In some embodiments, the status determination module 303 is configured to: judging a first running state of the equipment according to the current state data and the predicted value; clustering the historical state data and the current state data, and labeling an operating state label for each data cluster obtained by clustering; determining a second running state of the equipment according to the running state label of the data cluster to which the current state data belongs; the early warning decision module 304 is configured to: and sending a power failure early warning signal to the equipment under the condition that at least one of the first running state and the second running state is abnormal.
In some embodiments, the historical state data and the current state data comprise at least one of: historical operating voltage and current operating voltage; historical operating voltage and current operating current; historical operating power and current operating power; historical power consumption and current power consumption.
A specific application example of the early warning system for the operation state of the device of the present disclosure is described below with reference to fig. 4.
Fig. 4 shows a schematic structural diagram of a specific application example of the early warning system for the operation state of the device of the present disclosure. As shown in fig. 4, the early warning system 40 for the operation state of the device includes: the energy management system comprises an energy management subsystem 401, an intelligent GDC (Gateway data channel) Gateway 402, and four energy device state prediction devices (an energy device voltage prediction module 403, an energy device current prediction module 404, an energy device power prediction module 405, and an energy device power consumption prediction module 406). The individual energy device prediction means are designed individually in order to take into account the limited calculation power of the GDC. Each energy device state prediction means may include a model training switch and a model prediction switch. The energy device state predicting apparatus has an input and an output, and the energy device state predicting apparatus also has a reset function. When the model prediction switch is pressed, the energy device state prediction apparatus is ready to start operating, but before the energy device state prediction apparatus is formally operated, the model training function and the model prediction function need to be debugged. And when the switch is turned off or a reset key is pressed, the model is erased, and if the energy equipment state prediction device needs to be started again to stably operate, the energy equipment state prediction device needs to be debugged again. The energy equipment state predicting device also has a storage medium for storing the model, and the communication data is encrypted when the device operates.
The intelligent GDC gateway is in data connection with an underlying energy device (such as a photovoltaic system). The intelligent GDC gateway is composed of a data transmission module 4021, a judging device 4022 and a decision device 4023. The data transmission module 4021 is used for transmitting data of the bottom layer energy equipment; the evaluator 4022 is configured to evaluate the prediction result; the decision maker 4023 determines whether certain decisions need to be made according to the evaluation result.
The work flow of the whole system is as follows:
the state data of the bottom layer energy equipment collected by the sensor is transmitted to the data transmission module 4021 of the intelligent GDC gateway 402 through the CAN protocol, and then the data is transmitted to the database of the energy management subsystem 401 from the intelligent GDC gateway 402 through the TCP/IP protocol again for displaying the energy information data on a front-end page.
And (II) data are respectively input into the four energy equipment state prediction devices from the data transmission module 4021 and then are predicted, and the prediction results are respectively input into the judger 4022 after being output, and meanwhile, the intelligent GDC gateway 402 can also integrate the clustering algorithm module to directly predict the data. The judger 4022 makes a judgment according to the prediction result, and the decider 4023 determines how to decide according to the judgment result.
With respect to establishing the prediction model:
(1) and reading the energy equipment state data with the time continuity as long as possible in the database as a training sample of the prediction model.
(2) And preprocessing the training sample by using a data preprocessing method. For example, the null value is filled by using a data filling method based on attribute importance (reference: data filling method based on attribute importance (computer engineering and design)); firstly, identifying abnormal values by adopting a method based on a statistical theory, and completing the identified abnormal values by using a method based on attribute importance. Because the state data of the energy equipment has certain periodicity, the noise reduction can be carried out by adopting improved wavelet decomposition (reference: data filling method (traffic information and safety) based on attribute importance) to filter the interference of high-frequency noise.
(3) And decomposing the training sample after data preprocessing by using an LMD (local mean decomposition) method to obtain a plurality of PF (particle Filter) and residual components.
(4) And training and storing a corresponding Time-CNN model by using a plurality of PFs and residual components.
And (3) predicting the state data of the energy equipment:
(1) and (4) reading the energy equipment state data in the past week by a program between 23 o ' clock and 30 o ' clock and 24 o ' clock every day for predicting the energy equipment state.
(2) And preprocessing the state data of the energy equipment by using a data preprocessing method.
(3) And decomposing the energy equipment state data after data preprocessing by using an LMD method to obtain a plurality of PF and residual components.
(4) And loading the trained Time-CNN model into a memory from the hard disk, respectively bringing each PF component and residual component into the respective Time-CNN model, and outputting each PF component predicted value and residual component predicted value.
(5) And weighting and predicting the energy equipment state data of the future day by using each PF component predicted value and each residual component predicted value.
The application example realizes the early warning system of the running state of the equipment. The early warning system belongs to a distributed control system, adopts the edge calculation technology to carry out the calculation process in a layered way, and utilizes an energy equipment state prediction device to share part of calculation work of an intelligent GDC network manager.
Some embodiments of the warning device for the operation state of the apparatus of the present disclosure are described below with reference to fig. 5.
Fig. 5 is a schematic structural diagram of an early warning device for an operating state of an apparatus according to some embodiments of the present disclosure. As shown in fig. 5, the device 50 for warning the operation state of the equipment according to this embodiment includes: a memory 510 and a processor 520 coupled to the memory 510, the processor 520 being configured to perform the method of pre-warning of the device operating state in any of the embodiments described above based on instructions stored in the memory 510.
Memory 510 may include, for example, system memory, fixed non-volatile storage media, and the like. The system memory stores, for example, an operating system, an application program, a Boot Loader (Boot Loader), and other programs.
The early warning device 50 for the device operation state may further include an input/output interface 530, a network interface 540, a storage interface 550, and the like. These interfaces 530, 540, 550 and the connections between the memory 510 and the processor 520 may be, for example, via a bus 560. The input/output interface 530 provides a connection interface for input/output devices such as a display, a mouse, a keyboard, and a touch screen. The network interface 540 provides a connection interface for various networking devices. The storage interface 550 provides a connection interface for external storage devices such as an SD card and a usb disk.
The present disclosure also includes a computer-readable storage medium having stored thereon computer instructions that, when executed by a processor, implement a method for early warning of an operational state of a device in any of the foregoing embodiments.
The present disclosure is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the disclosure. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only exemplary of the present disclosure and is not intended to limit the present disclosure, so that any modification, equivalent replacement, or improvement made within the spirit and principle of the present disclosure should be included in the scope of the present disclosure.
Claims (20)
1. An early warning method for the running state of equipment comprises the following steps:
acquiring historical state data and current state data of equipment;
obtaining a predicted value of the current state data according to the historical state data;
judging the running state of the equipment according to the current state data and the predicted value;
and sending an early warning signal to the equipment according to the running state of the equipment.
2. The early warning method as claimed in claim 1, wherein the obtaining the predicted value of the current state data according to the historical state data comprises:
performing local mean decomposition on the historical state data in a time series form to obtain a plurality of PF components and a residual component of the historical state data;
inputting each PF component and residual component of the historical state data into a corresponding pre-trained time convolution neural network respectively to obtain the predicted values of each PF component and residual component of the current state data;
and determining the predicted value of the current state data according to the predicted values of each PF component and residual component of the current state data.
3. The warning method according to claim 2, wherein the determining the predicted value of the current state data according to the predicted values of the PF components and the residual components of the current state data comprises:
respectively determining cross-correlation coefficients between each PF component and residual component of the historical state data and the historical state data in a time series form;
and determining the predicted value of the current state data according to the cross correlation coefficient, the predicted values of each PF component and the residual component of the current state data.
4. A warning method as claimed in claim 3, wherein the determining the predicted value of the current state data from the cross-correlation coefficient, the predicted values of the PF components and the residual component of the current state data comprises: determining a predicted value of the current state data in the following manner
Wherein c represents the predicted value of the current state data, n represents the total number of PF components, i represents the identity of PF components, aiRepresenting the cross-correlation coefficient between the ith PF component and the historical state data in time series, PFiDenotes the ith PF component, b denotes the cross-correlation coefficient between the residual component and the historical state data in time series form, and S denotes the residual component.
5. The early warning method as claimed in claim 2, wherein the obtaining the predicted value of the current state data according to the historical state data further comprises:
performing local mean decomposition on training data in a time series form to obtain a plurality of PF components and a residual component of the training data;
and respectively training corresponding time convolution neural networks by using each PF component and residual component of the training data, so that each trained time convolution neural network can output a predicted value of the PF component according to the input PF component or output a predicted value of the residual component according to the input residual component.
6. The warning method as claimed in claim 1, wherein,
the acquiring historical state data and current state data of the device comprises: periodically acquiring historical state data and current state data of equipment;
the obtaining a predicted value of the current state data according to the historical state data comprises: periodically obtaining a predicted value of the current state data according to the historical state data;
the judging the running state of the equipment according to the current state data and the predicted value comprises the following steps: periodically determining a deviation between the current state data and the predicted value; and under the condition that the frequency of the deviation larger than the first threshold is larger than the second threshold, judging that the running state of the equipment is abnormal.
7. The warning method as claimed in claim 6, wherein the transmitting the warning signal to the device according to the operation state of the device comprises:
and sending a power failure early warning signal to the equipment under the condition that the running state of the equipment is abnormal.
8. The warning method as claimed in claim 1, wherein,
the judging the running state of the equipment according to the current state data and the predicted value comprises the following steps: judging a first running state of the equipment according to the current state data and the predicted value; clustering the historical state data and the current state data, and labeling an operating state label for each data cluster obtained by clustering; determining a second running state of the equipment according to the running state label of the data cluster to which the current state data belongs;
the sending of the early warning signal to the device according to the running state of the device comprises: and sending a power failure early warning signal to equipment under the condition that at least one of the first running state and the second running state is abnormal.
9. The warning method of any one of claims 1 to 8, wherein the historical and current status data comprises at least one of:
historical operating voltage and current operating voltage;
historical operating voltage and current operating current;
historical operating power and current operating power;
historical power consumption and current power consumption.
10. An early warning system for the operating state of a device, comprising:
the data transmission module is configured to acquire historical state data and current state data of the equipment;
a data prediction module configured to obtain a prediction value of the current state data according to the historical state data;
the state judgment module is configured to judge the running state of the equipment according to the current state data and the predicted value;
and the early warning decision module is configured to send an early warning signal to the equipment according to the running state of the equipment.
11. The warning system of claim 10, wherein the data prediction module is configured to:
performing local mean decomposition on the historical state data in a time series form to obtain a plurality of PF components and a residual component of the historical state data;
inputting each PF component and residual component of the historical state data into a corresponding pre-trained time convolution neural network respectively to obtain the predicted values of each PF component and residual component of the current state data;
and determining the predicted value of the current state data according to the predicted values of each PF component and residual component of the current state data.
12. The warning system of claim 11, wherein the data prediction module is configured to:
respectively determining cross-correlation coefficients between each PF component and residual component of the historical state data and the historical state data in a time series form;
and determining the predicted value of the current state data according to the cross correlation coefficient, the predicted values of each PF component and the residual component of the current state data.
13. The warning system of claim 12, wherein the data prediction module is configured to: determining a predicted value of the current state data in the following manner
Wherein c represents the predicted value of the current state data, n represents the total number of PF components, i represents the identity of PF components, aiRepresenting the cross-correlation coefficient between the ith PF component and the historical state data in time series, PFiDenotes the ith PF component, b denotes the cross-correlation coefficient between the residual component and the historical state data in time series form, and S denotes the residual component.
14. The warning system of claim 11, wherein the data prediction module is further configured to:
performing local mean decomposition on training data in a time series form to obtain a plurality of PF components and a residual component of the training data;
and respectively training corresponding time convolution neural networks by using each PF component and residual component of the training data, so that each trained time convolution neural network can output a predicted value of the PF component according to the input PF component or output a predicted value of the residual component according to the input residual component.
15. The warning system as claimed in claim 10, wherein,
the data transmission module is configured to: periodically acquiring historical state data and current state data of equipment;
the data prediction module is configured to: periodically obtaining a predicted value of the current state data according to the historical state data;
the state determination module is configured to: periodically determining a deviation between the current state data and the predicted value; and under the condition that the frequency of the deviation larger than the first threshold is larger than the second threshold, judging that the running state of the equipment is abnormal.
16. The early warning system of claim 15, wherein the early warning decision module is configured to:
and sending a power failure early warning signal to the equipment under the condition that the running state of the equipment is abnormal.
17. The warning system as claimed in claim 10, wherein,
the state determination module is configured to: judging a first running state of the equipment according to the current state data and the predicted value; clustering the historical state data and the current state data, and labeling an operating state label for each data cluster obtained by clustering; determining a second running state of the equipment according to the running state label of the data cluster to which the current state data belongs;
the early warning decision module is configured to: and sending a power failure early warning signal to equipment under the condition that at least one of the first running state and the second running state is abnormal.
18. An early warning system as claimed in any one of claims 10 to 17, wherein the historical and current status data includes at least one of:
historical operating voltage and current operating voltage;
historical operating voltage and current operating current;
historical operating power and current operating power;
historical power consumption and current power consumption.
19. An early warning device for the running state of equipment comprises:
a memory; and
a processor coupled to the memory, the processor configured to execute the method of pre-warning of operational status of a device according to any one of claims 1 to 9 based on instructions stored in the memory.
20. A computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions which, when executed by a processor, implement the method for warning of the operational state of a device according to any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910898842.6A CN110658905B (en) | 2019-09-23 | 2019-09-23 | Early warning method, early warning system and early warning device for equipment operation state |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910898842.6A CN110658905B (en) | 2019-09-23 | 2019-09-23 | Early warning method, early warning system and early warning device for equipment operation state |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110658905A true CN110658905A (en) | 2020-01-07 |
| CN110658905B CN110658905B (en) | 2023-08-04 |
Family
ID=69038846
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910898842.6A Active CN110658905B (en) | 2019-09-23 | 2019-09-23 | Early warning method, early warning system and early warning device for equipment operation state |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110658905B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111275185A (en) * | 2020-01-16 | 2020-06-12 | 珠海格力电器股份有限公司 | Energy use state early warning method, device, equipment and storage medium |
| CN112034387A (en) * | 2020-09-08 | 2020-12-04 | 武汉大学 | Power transmission line short-circuit fault diagnosis method and device based on prediction sequence |
| CN113923096A (en) * | 2020-06-22 | 2022-01-11 | 中国联合网络通信集团有限公司 | Network element fault early warning method and device, electronic equipment and storage medium |
| CN117273547A (en) * | 2023-11-17 | 2023-12-22 | 建平慧营化工有限公司 | Production equipment operation data processing method based on edge calculation |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100262411A1 (en) * | 2006-02-14 | 2010-10-14 | Edsa Micro Corporation | Systems and methods for real-time system monitoring and predictive analysis |
| CN106875041A (en) * | 2017-01-16 | 2017-06-20 | 广东电网有限责任公司揭阳供电局 | A kind of short-term wind speed forecasting method |
| CN107608862A (en) * | 2017-10-13 | 2018-01-19 | 众安信息技术服务有限公司 | Monitoring alarm method, monitoring alarm device and computer-readable recording medium |
| CN107832896A (en) * | 2017-11-29 | 2018-03-23 | 广东电网有限责任公司电力科学研究院 | A kind of electric power factory equipment soft fault method for early warning and device |
| CN109242532A (en) * | 2018-08-03 | 2019-01-18 | 广东工业大学 | The Short-term electricity price forecasting method of RBF neural is decomposed and optimized based on local mean value |
| CN109639485A (en) * | 2018-12-13 | 2019-04-16 | 国家电网有限公司 | The monitoring method and device of electricity consumption acquisition communication link |
-
2019
- 2019-09-23 CN CN201910898842.6A patent/CN110658905B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100262411A1 (en) * | 2006-02-14 | 2010-10-14 | Edsa Micro Corporation | Systems and methods for real-time system monitoring and predictive analysis |
| CN106875041A (en) * | 2017-01-16 | 2017-06-20 | 广东电网有限责任公司揭阳供电局 | A kind of short-term wind speed forecasting method |
| CN107608862A (en) * | 2017-10-13 | 2018-01-19 | 众安信息技术服务有限公司 | Monitoring alarm method, monitoring alarm device and computer-readable recording medium |
| CN107832896A (en) * | 2017-11-29 | 2018-03-23 | 广东电网有限责任公司电力科学研究院 | A kind of electric power factory equipment soft fault method for early warning and device |
| CN109242532A (en) * | 2018-08-03 | 2019-01-18 | 广东工业大学 | The Short-term electricity price forecasting method of RBF neural is decomposed and optimized based on local mean value |
| CN109639485A (en) * | 2018-12-13 | 2019-04-16 | 国家电网有限公司 | The monitoring method and device of electricity consumption acquisition communication link |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111275185A (en) * | 2020-01-16 | 2020-06-12 | 珠海格力电器股份有限公司 | Energy use state early warning method, device, equipment and storage medium |
| CN113923096A (en) * | 2020-06-22 | 2022-01-11 | 中国联合网络通信集团有限公司 | Network element fault early warning method and device, electronic equipment and storage medium |
| CN113923096B (en) * | 2020-06-22 | 2023-05-30 | 中国联合网络通信集团有限公司 | Network element fault early warning method and device, electronic equipment and storage medium |
| CN112034387A (en) * | 2020-09-08 | 2020-12-04 | 武汉大学 | Power transmission line short-circuit fault diagnosis method and device based on prediction sequence |
| CN112034387B (en) * | 2020-09-08 | 2021-09-21 | 武汉大学 | Power transmission line short-circuit fault diagnosis method and device based on prediction sequence |
| CN117273547A (en) * | 2023-11-17 | 2023-12-22 | 建平慧营化工有限公司 | Production equipment operation data processing method based on edge calculation |
| CN117273547B (en) * | 2023-11-17 | 2024-01-30 | 建平慧营化工有限公司 | Production equipment operation data processing method based on edge calculation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110658905B (en) | 2023-08-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110658905A (en) | Early warning method, early warning system and early warning device for equipment running state | |
| CN111241154B (en) | Storage battery fault early warning method and system based on big data | |
| CN112069795A (en) | Corpus detection method, apparatus, device and medium based on mask language model | |
| CN116843071B (en) | Transportation network operation index prediction method and device for intelligent port | |
| CN113435998A (en) | Loan overdue prediction method and device, electronic equipment and storage medium | |
| CN115099326A (en) | Behavior prediction method, behavior prediction device, behavior prediction equipment and storage medium based on artificial intelligence | |
| CN111007452A (en) | Fault diagnosis method and device of data acquisition system | |
| CN114816468A (en) | Cloud edge coordination system, data processing method, electronic device and storage medium | |
| CN112070180B (en) | Power grid equipment state judging method and device based on information physical bilateral data | |
| CN110413482B (en) | Detection method and device | |
| CN110826867A (en) | Vehicle management method, device, computer equipment and storage medium | |
| CN105227410A (en) | Based on the method and system that the server load of adaptive neural network detects | |
| CN115577927A (en) | Important power consumer electricity utilization safety assessment method and device based on rough set | |
| CN115203014A (en) | Ecological service abnormity restoration system and restoration method based on deep learning | |
| US8468107B2 (en) | Non-intrusive event-driven prediction | |
| CN114968336A (en) | Application gray level publishing method and device, computer equipment and storage medium | |
| CN112710979A (en) | Intelligent electric energy meter operation monitoring management system and method based on deep learning | |
| CN114089206A (en) | Method, system, medium and device for predicting service life of battery redundancy module | |
| CN112800037A (en) | Optimization method and device for engineering cost data processing | |
| Esteves et al. | Prognostics health management: perspectives in engineering systems reliability prognostics | |
| Praczyk et al. | Finding relevant process characteristics with a method for data-based complexity reduction | |
| CN111143774A (en) | Power load prediction method and device based on influence factor multi-state model | |
| Vachtsevanos et al. | Prognosis: Challenges, Precepts, Myths and Applications | |
| CN115841255B (en) | On-site early warning method and system for building engineering based on-line analysis | |
| CN113807990A (en) | Construction site safety training method, system, device and storage medium |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |