Power plant equipment state maintenance decision-making method and system based on big data analysis

Abstract

The invention discloses a power plant equipment state maintenance decision-making method and system based on big data analysis, and relates to the technical field of equipment state prediction, and the method comprises the steps: collecting the operation data of power plant equipment in real time; extracting operation equipment characteristics, performing equipment state analysis, and predicting faults; and providing maintenance decision support based on equipment fault prediction. According to the power plant equipment state maintenance decision-making method based on big data analysis, feature fusion is carried out on the extracted vibration features and pressure features, the health scores of the equipment are calculated, whether the equipment in the low health state will break down in the future time period or not is determined, redundant maintenance is avoided, and the maintenance efficiency is improved. The maintenance efficiency is remarkably improved, the cost is reduced, specific maintenance measures in different states are provided according to the equipment state predicted value, the unnecessary maintenance and downtime is reduced, and the equipment management efficiency is improved. According to the invention, better effects are achieved in the aspects of reducing the cost and improving the management efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment status prediction, and in particular to a method and system for power plant equipment status maintenance decision-making based on big data analysis. Background Art

[0002] With the rapid development of modern industry, power plants are important energy production institutions, and the stable operation of their equipment is crucial to ensuring power supply. In the past few decades, the monitoring and maintenance technology of power plant equipment has undergone a transformation from traditional periodic maintenance to condition-based maintenance (CBM). The traditional periodic maintenance method mainly relies on the operation time or use cycle of the equipment for maintenance. Although it is simple and easy, it is easy to cause problems such as over-maintenance or under-maintenance, which wastes resources and fails to detect potential fault hazards in time. With the advancement of sensor technology, data acquisition and processing technology, condition-based maintenance strategies have gradually become mainstream. Condition monitoring technology collects various parameters of equipment operation in real time, such as vibration, temperature, pressure, etc., and combines data analysis technology to evaluate the equipment status, thereby achieving more accurate fault prediction and preventive maintenance.

[0003] However, existing condition monitoring and maintenance decision-making methods still have some shortcomings. First, traditional condition monitoring systems rely on single or a few sensor data, with limited data sources, making it difficult to fully reflect the operating status of the equipment, resulting in insufficient prediction accuracy. Second, existing methods have limited capabilities in data processing and feature extraction, especially when processing nonlinear and non-stationary signals. In addition, many existing fault prediction models lack adaptability to the complex and changeable equipment operating environment, and are unable to adjust and optimize maintenance strategies in a timely manner. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is that the existing periodic maintenance method has the problems of over-maintenance or under-maintenance, low maintenance efficiency, difficulty in accurately predicting equipment failures and inability to adjust maintenance strategies in a timely manner.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a power plant equipment condition maintenance decision method based on big data analysis, including real-time collection of power plant equipment operation data, extraction of operating equipment characteristics, equipment condition analysis, and fault prediction, and providing maintenance decision support based on equipment fault prediction.

[0007] As a preferred solution of the power plant equipment status maintenance decision-making method based on big data analysis described in the present invention, the real-time collection of power plant equipment operation data includes configuring vibration sensors and pressure sensors, collecting vibration data x(t) and pressure data P(t) in real time, and normalizing the data.

[0008] As a preferred solution of the power plant equipment condition maintenance decision-making method based on big data analysis described in the present invention, wherein: the extraction of operating equipment features includes extracting features from the collected vibration data through wavelet transform,

[0009]

[0010] Among them, a is the scale, b is the translation parameter, and ψ is the mother wavelet function.

[0011] The nonlinear trend of pressure data change is approximated by high-order terms for feature extraction, which can be expressed as:

[0012]

[0013] Among them, c i is the polynomial fitting coefficient, k j is the Gaussian kernel coefficient, μ j and σ j are the mean and standard deviation of the Gaussian kernel.

[0014] As a preferred solution of the power plant equipment condition maintenance decision method based on big data analysis described in the present invention, the equipment condition analysis includes fusing the extracted vibration features and pressure features, which is expressed as:

[0015]

[0016] Among them, α1, α2 and α3 are fusion weight parameters, β is the exponential function parameter, W x (a i ,b i 0 is the wavelet transform feature of vibration data, c i,j is the polynomial fitting coefficient of pressure data, P i It is the pressure data feature.

[0017] After feature fusion is completed, equipment status health analysis is performed based on the fused features.

[0018] As a preferred solution of the power plant equipment condition maintenance decision-making method based on big data analysis described in the present invention, the equipment condition health analysis includes inputting the extracted feature vector into the health assessment model for evaluation, which is expressed as:

[0019]

[0020] Among them, α i is the weight of the support vector, γ is the Gaussian kernel function parameter, and b is the bias term. The health status score H(t) of the device at different time points is obtained. The higher the score, the healthier the device.

[0021] As a preferred solution of the power plant equipment condition maintenance decision method based on big data analysis described in the present invention, wherein: the provision of maintenance decision support based on equipment fault prediction includes, when the equipment health score does not reach the preset health value, it is considered that the equipment may fail within the future time Δt that matches the preset health threshold, and the equipment is predicted to fail, and the interval vector features of the historical time points are extracted, which are expressed as:

[0022] F(t-iΔt)=[f(t-Δt),f(t-2Δt),…,f(t-nΔt)]

[0023] Among them, Δt is the time interval, n is the number of time steps, and the equipment status at future time points is predicted through time series modeling, which is expressed as:

[0024]

[0025] in, is the predicted value of the device state at the future time point t+Δt, w(t) is the feature vector weight at time t, ⊙ represents the element-by-element product, and the feature vector weight is expressed as:

[0026]

[0027] Determine the maintenance strategy of the equipment based on the predicted status value.

[0028] As a preferred solution of the power plant equipment status maintenance decision-making method based on big data analysis described in the present invention, wherein: the provision of maintenance decision support based on equipment failure prediction also includes that when the equipment status prediction value at the future time point t+Δt is ≥0.8, the equipment status is good and monitoring continues; when 0.5≤the equipment status prediction value at the future time point t+Δt<0.8, the equipment status is slightly abnormal, the monitoring frequency is increased, and a maintenance warning is issued to the operation and maintenance personnel to determine whether maintenance is required; when the equipment status prediction value at the future time point t+Δt is <0.5, the equipment needs to be currently maintained; if no maintenance is performed, the equipment is marked as high-risk and awaiting maintenance and is stopped from use.

[0029] Another object of the present invention is to provide a power plant equipment status maintenance decision system based on big data analysis, which can fuse the extracted vibration features and pressure features and calculate the health score of the equipment to determine whether the equipment with poor health status will fail in the future time period, thereby solving the problem of excessive maintenance contained in the current periodic inspection technology.

[0030] As a preferred solution of the power plant equipment status maintenance decision system based on big data analysis described in the present invention, it includes: a data acquisition module, a status analysis module, and a decision support module. The data acquisition module is used to collect power plant equipment operation data in real time. The status analysis module is used to extract the characteristics of the operating equipment, perform equipment status analysis, and predict faults. The decision support module is used to provide maintenance decision support based on equipment fault prediction.

[0031] A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement a step of a power plant equipment status maintenance decision-making method based on big data analysis.

[0032] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a power plant equipment condition maintenance decision-making method based on big data analysis.

[0033] Beneficial effects of the present invention: The power plant equipment condition maintenance decision method based on big data analysis provided by the present invention performs feature fusion of the extracted vibration features and pressure features, and calculates the health score of the equipment to determine whether the equipment with low health status will fail in the future time period, thereby avoiding unnecessary maintenance, significantly improving maintenance efficiency and reducing costs, and providing specific maintenance measures under different states according to the equipment status prediction value, reducing unnecessary maintenance and downtime, and improving equipment management efficiency. The present invention achieves better results in reducing costs and improving management efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0035] Figure 1 An overall flow chart of a power plant equipment condition maintenance decision-making method based on big data analysis provided in the first embodiment of the present invention.

[0036] Figure 2A module diagram of a power plant equipment status maintenance decision system based on big data analysis provided in the third embodiment of the present invention. DETAILED DESCRIPTION

[0037] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0038] Example 1

[0039] Reference Figure 1 , which is an embodiment of the present invention, provides a power plant equipment condition maintenance decision method based on big data analysis, comprising:

[0040] S1: Real-time collection of power plant equipment operation data.

[0041] Furthermore, real-time collection of power plant equipment operation data includes configuring vibration sensors and pressure sensors, collecting vibration data x(t) and pressure data P(t) in real time, and normalizing the data.

[0042] S2: Extract the characteristics of running equipment, analyze the equipment status, and predict failures.

[0043] Furthermore, extracting the characteristics of the operating equipment includes extracting the characteristics of the collected vibration data through wavelet transform.

[0044]

[0045] Among them, a is the scale, b is the translation parameter, As the mother wavelet function, vibration data usually has complex nonlinear and non-stationary characteristics. Traditional time domain and frequency domain analysis methods are difficult to effectively extract characteristic information from vibration signals. As a time-frequency analysis tool, wavelet transform can provide time and frequency information at the same time and has good analytical ability for the local characteristics of vibration signals.

[0046] The nonlinear trend of pressure data change is approximated by high-order terms for feature extraction, which can be expressed as:

[0047]

[0048] Among them, c i is the polynomial fitting coefficient, k j is the Gaussian kernel coefficient, μ j and σ jare the mean and standard deviation of the Gaussian kernel.

[0049] It should be noted that the equipment status analysis includes the feature fusion of the extracted vibration features and pressure features, which can be expressed as:

[0050]

[0051] Among them, α1, α2 and α3 are fusion weight parameters, β is the exponential function parameter, W x (a i ,b i ) is the wavelet transform feature of vibration data, c i,j is the polynomial fitting coefficient of pressure data, P i It is the pressure data feature.

[0052] After feature fusion is completed, equipment status health analysis is performed based on the fused features.

[0053] S3: Provide maintenance decision support based on equipment failure prediction.

[0054] Furthermore, maintenance decision support is provided based on equipment failure prediction, including that when the equipment health score does not reach the preset health value, the equipment is considered to have a possibility of failure within the future time Δt that matches the preset health threshold, and the equipment is predicted to fail, and the interval vector features of the historical time points are extracted, which are expressed as:

[0055] F(t-iΔt)=[f(t-Δt),f(t-2Δt),…,f(t-nΔt)]

[0056] Among them, Δt is the time interval, n is the number of time steps, and the equipment status at future time points is predicted through time series modeling, which is expressed as:

[0057]

[0058] in, is the predicted value of the device state at the future time point t+Δt, w(t) is the feature vector weight at time t, ⊙ represents the element-by-element product, and the feature vector weight is expressed as:

[0059]

[0060] Determine the maintenance strategy of the equipment based on the predicted status value.

[0061] It should be noted that, first, by real-time monitoring of the equipment health score, when the score does not reach the preset health value, the system identifies that the equipment may have potential failure risks. Subsequently, by extracting the interval vector features of historical time points, building a time series model, predicting the future equipment status, and calculating the status prediction value at future time points, the time value setting here is related to the health value threshold. The higher the health value threshold is set, the longer the interval from the current time to the future time, because the increase in the health threshold can improve the system's longer-term health protection for the equipment, but it is not the case that the higher the health value, the better, usually set at 80% or 60%. According to the prediction results, the system can automatically generate maintenance strategies for different states and provide scientific maintenance decision support.

[0062] Specifically, when the equipment health score does not reach the preset value, the system identifies the possibility of equipment failure and extracts the interval vector features of historical time points to perform time series modeling. By calculating the predicted value of the equipment status at future time points, the system can provide early warning and generate corresponding maintenance strategies. These steps are closely combined to form a complete closed-loop system, which effectively improves the accuracy of equipment status monitoring and fault prediction, optimizes maintenance strategies, reduces equipment maintenance costs, and improves the reliability and safety of equipment operation.

[0063] It should be noted that providing maintenance decision support based on equipment failure prediction also includes that when the predicted value of the equipment status at the future time point t+Δt is ≥0.8, the equipment status is good and monitoring continues; when 0.5≤the predicted value of the equipment status at the future time point t+Δt<0.8, the equipment status is slightly abnormal, the monitoring frequency is increased, and a maintenance warning is issued to the operation and maintenance personnel to determine whether maintenance is required; when the predicted value of the equipment status at the future time point t+Δt is <0.5, the equipment needs to be repaired and maintained currently. If no maintenance is performed, the equipment is marked as high-risk and awaiting maintenance and is stopped from use.

[0064] Example 2

[0065] An embodiment of the present invention provides a power plant equipment condition maintenance decision-making method based on big data analysis. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0066] First, in order to verify the effectiveness of the present invention, the following experiments were conducted. The experiments were conducted on a steam turbine actually operating in a power plant. The main reason for selecting this equipment is its complex operating state and frequent failure requirements, which can fully demonstrate the superiority of the invention method. The experiments mainly include four stages: data collection, feature extraction, state analysis, and maintenance decision-making.

[0067] Vibration sensors and pressure sensors are configured to collect vibration data and pressure data of the equipment in real time. The collected data is normalized through the data acquisition system to ensure data standardization and consistency.

[0068] The vibration data is characterized by wavelet transform to capture the mutation and periodic characteristics in the vibration signal.

[0069] High-order polynomial fitting and Gaussian kernel function approximation are used to extract the nonlinear variation characteristics of the pressure data.

[0070] The extracted vibration features and pressure features are fused, and the importance of each feature is synthesized in a weighted manner.

[0071] The fused feature vectors were input into the health assessment model to calculate the health status score of the equipment, with 0.6 as the health threshold. During the test, real-time data collection and analysis were performed for 100 consecutive days, and data collection and evaluation were performed every ten days to record the vibration and pressure characteristics of the equipment at different time points and the corresponding health scores.

[0072] Table 1 Experimental data table

[0073]

[0074]

[0075] From the changing trend of the health score, it can be seen that with the passage of time, the health score of the equipment gradually decreases, reflecting the deterioration of the equipment status. This is consistent with the actual operation of the equipment, which verifies the accuracy of the invented method in condition monitoring. The method of the present invention predicts the future state of the equipment through feature fusion and time series modeling, so that maintenance strategies can be formulated in advance. For example, at time point T8, the health score is 0.50, and the predicted value of the equipment status shows that the equipment may fail in the next 40 days. The system issues a warning signal and recommends equipment maintenance. On the 90th-100th day without maintenance, the status of the equipment has turned abnormal. It can be seen that the maintenance decision support for the equipment is reliable.

[0076] Example 3

[0077] Reference Figure 2 , is an embodiment of the present invention, which provides a power plant equipment condition maintenance decision system based on big data analysis, including a data acquisition module, a condition analysis module, and a decision support module;

[0078] Among them, the data acquisition module is used to collect power plant equipment operation data in real time; the status analysis module is used to extract the characteristics of operating equipment, perform equipment status analysis, and predict faults; and the decision support module is used to provide maintenance decision support based on equipment fault prediction.

[0079] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0080] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0081] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0082] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc. It should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and are not limited. Although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

[0083] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A power plant equipment condition maintenance decision-making method based on big data analysis, characterized in that: include: Real-time collection of power plant equipment operation data; Extract the characteristics of running equipment, analyze equipment status, and predict failures; Provide maintenance decision support based on equipment failure prediction.

2. The power plant equipment condition maintenance decision-making method based on big data analysis according to claim 1 is characterized by: The real-time collection of power plant equipment operation data includes configuring a vibration sensor and a pressure sensor, collecting vibration data x(t) and pressure data P(t) in real time, and normalizing the data.

3. The power plant equipment condition maintenance decision-making method based on big data analysis according to claim 2 is characterized by: The extracting of the running equipment features includes extracting features from the collected vibration data by wavelet transform. Among them, a is the scale, b is the translation parameter, and ψ is the mother wavelet function; The nonlinear trend of pressure data change is approximated by high-order terms for feature extraction, which can be expressed as: Among them, c i is the polynomial fitting coefficient, k j is the Gaussian kernel coefficient, μ j and σ j are the mean and standard deviation of the Gaussian kernel.

4. The power plant equipment condition maintenance decision-making method based on big data analysis according to claim 3 is characterized by: The equipment status analysis includes fusing the extracted vibration features and pressure features, which can be expressed as: Among them, α1, α2 and α3 are fusion weight parameters, β is the exponential function parameter, W x (a i ,b i ) is the wavelet transform feature of vibration data, c i,j is the polynomial fitting coefficient of pressure data, P i is the pressure data feature; After feature fusion is completed, equipment status health analysis is performed based on the fused features.

5. The power plant equipment condition maintenance decision-making method based on big data analysis according to claim 4 is characterized by: The equipment status health analysis includes inputting the extracted feature vector into the health assessment model for evaluation, which is expressed as: Among them, α i is the weight of the support vector, γ is the Gaussian kernel function parameter, and b is the bias term. The health status score H(t) of the device at different time points is obtained. The higher the score, the healthier the device.

6. The power plant equipment condition maintenance decision-making method based on big data analysis according to claim 5 is characterized by: The maintenance decision support based on equipment failure prediction includes: when the equipment health score does not reach the preset health value, it is considered that the equipment may fail within the future time Δt that matches the preset health threshold, and the equipment is predicted to fail, and the interval vector features of the historical time point are extracted, which are expressed as: F(t-iΔt)=[f(t-Δt),f(t-2Δt),…,f(t-nΔt)] Among them, Δt is the time interval, n is the number of time steps, and the equipment status at future time points is predicted through time series modeling, which is expressed as: in, is the predicted value of the device state at the future time point t+Δt, w(t) is the feature vector weight at time t, ⊙ represents the element-by-element product, and the feature vector weight is expressed as: Determine the maintenance strategy of the equipment based on the predicted status value.

7. The power plant equipment condition maintenance decision-making method based on big data analysis according to claim 6 is characterized by: The maintenance decision support based on equipment failure prediction also includes that when the predicted value of the equipment state at the future time point t+Δt is ≥0.8, the equipment state is good and monitoring continues; when 0.5≤the predicted value of the equipment state at the future time point t+Δt<0.8, the equipment state is slightly abnormal, the monitoring frequency is increased, and a maintenance warning is issued to the operation and maintenance personnel to determine whether maintenance is required; when the predicted value of the equipment state at the future time point t+Δt is <0.5, the equipment needs to be currently maintained; if maintenance is not performed, the equipment is marked as high-risk and awaiting maintenance and is stopped from use.

8. A system using the power plant equipment condition maintenance decision-making method based on big data analysis as claimed in any one of claims 1 to 7, characterized in that: Including data acquisition module, status analysis module, and decision support module; The data acquisition module is used to collect power plant equipment operation data in real time; The state analysis module is used to extract the characteristics of the running equipment, perform equipment state analysis, and predict faults; The decision support module is used to provide maintenance decision support based on equipment failure prediction.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the power plant equipment condition maintenance decision-making method based on big data analysis described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the power plant equipment condition maintenance decision-making method based on big data analysis described in any one of claims 1 to 7 are implemented.

Images