CN114519923A - Intelligent diagnosis and early warning method and system for power plant - Google Patents
Intelligent diagnosis and early warning method and system for power plant Download PDFInfo
- Publication number
- CN114519923A CN114519923A CN202111408475.0A CN202111408475A CN114519923A CN 114519923 A CN114519923 A CN 114519923A CN 202111408475 A CN202111408475 A CN 202111408475A CN 114519923 A CN114519923 A CN 114519923A
- Authority
- CN
- China
- Prior art keywords
- data
- power plant
- early warning
- model
- diagnosis
- 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
- 238000000034 method Methods 0.000 title claims abstract description 59
- 238000003745 diagnosis Methods 0.000 title claims abstract description 53
- 238000013499 data model Methods 0.000 claims abstract description 22
- 239000011159 matrix material Substances 0.000 claims abstract description 20
- 238000004458 analytical method Methods 0.000 claims abstract description 15
- 238000012544 monitoring process Methods 0.000 claims abstract description 14
- 238000011156 evaluation Methods 0.000 claims description 32
- 238000001514 detection method Methods 0.000 claims description 28
- 230000006870 function Effects 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 9
- 238000012360 testing method Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000001914 filtration Methods 0.000 claims description 4
- 230000007787 long-term memory Effects 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 4
- 238000009795 derivation Methods 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 3
- 230000001360 synchronised effect Effects 0.000 claims description 3
- 238000012800 visualization Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 2
- 238000006731 degradation reaction Methods 0.000 abstract description 6
- 230000003993 interaction Effects 0.000 abstract description 6
- 230000000007 visual effect Effects 0.000 abstract description 6
- 230000015556 catabolic process Effects 0.000 abstract description 5
- 238000004422 calculation algorithm Methods 0.000 abstract description 3
- 230000035784 germination Effects 0.000 abstract description 2
- 238000007405 data analysis Methods 0.000 description 11
- 238000012423 maintenance Methods 0.000 description 11
- 230000015654 memory Effects 0.000 description 11
- 238000005516 engineering process Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 3
- 238000013178 mathematical model Methods 0.000 description 3
- 238000010248 power generation Methods 0.000 description 3
- 238000009825 accumulation Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 239000006185 dispersion Substances 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- 230000008439 repair process Effects 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000002405 diagnostic procedure Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012067 mathematical method Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 239000000779 smoke Substances 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- -1 wind Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/18—Status alarms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention discloses an intelligent diagnosis and early warning method and system for a power plant, belonging to the field of state monitoring and fault diagnosis. The data access module realizes reading operation with a power plant database; the diagnosis and early warning analysis module builds a data model according to historical data, then diagnoses early fault germination by utilizing a robust nonnegative matrix factorization algorithm based on model output and real-time data, determines a failure threshold value by combining the running performance of the system, and realizes prediction of the complete failure time of the system based on Gamma degradation modeling; the visual interaction module is used for displaying the current monitoring state of the system and the future fault trend.
Description
Technical Field
The invention belongs to the field of state monitoring and fault diagnosis, and relates to an intelligent diagnosis and early warning method and system for a power plant.
Background
The equipment of the thermal power plant can be basically divided into a mechanical, furnace, electric and instrument 4 major industry series, mainly comprising a boiler, a steam turbine, a generator, a transformer, a large pump, a small pump, a fan, an electric switch, pipelines such as steam, water, wind, smoke, oil and pressure vessels, and the corresponding fault types are mainly mechanical faults and electrical faults. The system can be divided into a series of complex multi-coupling thermotechnical control loops of coordinated control, AGC control, primary air control, water supply control, overheat air temperature control and the like according to the system, and corresponding faults comprise actuator faults, sensor faults, control performance degradation and the like. At present, most thermal power plants still adopt the strategies of planned maintenance and fault maintenance, so that economic resources of the power plants are wasted, and meanwhile, the equipment is not maintained enough or maintained excessively, so that the health management of the equipment is not facilitated. Therefore, economic losses caused by equipment failure are increasingly paid attention and paid to power generation enterprises. In this case, the operation and maintenance mode of monitoring the state before the equipment failure, i.e. predicting the occurrence of the possible failure in advance, must bring wide attention to the power plant.
Conventional condition monitoring and fault diagnosis methods are generally classified into three categories, namely analytical model-based methods, knowledge-based methods, and data analysis-based methods. In 1971, the board doctor of the national institute of technology and technology of Massachusetts proposed a redundancy analysis technology and replaced the hardware redundancy technology, which lays the theoretical foundation for fault diagnosis. In 1976, a review of fault diagnosis was published by Willsky in Automatica, which was considered the earliest article on fault diagnosis. In 1978, himmelbau written the first academic book about the contents of fault diagnosis, and the method of fault diagnosis using the analytical model was widely used. The method based on the analytic model utilizes the prior information of the system performance parameters contained in the dynamic mathematical model, namely the invariable analytic relation between the input and the output of the system, compares the actual output value of the system with the expected output value to obtain a residual error, and utilizes the residual error to detect and identify the state and the fault of the residual error. The fault diagnosis method based on the analytical model is generally suitable for simple systems or processes with relatively small input/output and state numbers, and in the case, a mathematical model with higher precision can be obtained, so that the fault type can be detected more accurately. However, for a complex control system, characteristics such as nonlinearity, coupling, time-varying property and the like exist among variables, and an effective and accurate mathematical model is difficult to build. In addition, since the actual control system is influenced by uncertain factors such as noise, external interference and the like, the method is also disabled, and the application of the method is limited. Knowledge-based methods are divided into fault tree analysis, expert systems, symbolic directed graphs, neural networks, and the like. The Fault Tree Analysis (Fault Tree Analysis) is from the system terminal Fault, from top to bottom, layer by layer, so as to find out the most fundamental Fault factor and express the relation between the events and the system Fault in the form of logic diagram. Such methods are suitable for systems with a large amount of knowledge. However, the current knowledge processing method has certain difficulties, such as bottleneck problem of knowledge acquisition, limitation of knowledge reasoning, poor self-learning capability and self-adaptive capability, and the like.
The method based on data analysis is used for mining the implicit information in the data through various data processing and analyzing methods such as statistical analysis, cluster analysis, spectrum analysis, wavelet analysis and the like. The main advantage of the data analysis method is that a large amount of relevant data of the control system is utilized, effective information is extracted through a mathematical method, and useful statistical data and computational reasoning information are provided for field workers to improve the monitoring performance of the system. Data analysis-based methods project the dominant features into a low-dimensional space through rigorous statistical analysis, enabling a great simplification and improvement of the diagnostic process. Therefore, the fault diagnosis method based on data analysis has great theoretical significance and practical application value. With the rapid development of information technology, particularly the large-scale construction of DCS system, SIS system and various information systems, the data quantity accumulated by the power plant is increased rapidly, and a foundation is laid for the large data analysis and research work of key equipment of power generation enterprises. Through a special and efficient real-time data mining technology, a new generation of equipment state online monitoring and fault early warning system can help a user to realize intelligent management of equipment states, further give full play to the professional efficiency of an equipment manager, change fault post-treatment into pre-prevention, master the whole dynamic change of the equipment in operation in real time, greatly improve the operation safety level and efficiency of the equipment in the life cycle of each equipment, reduce unplanned shutdown and accidents caused by equipment reasons, reduce the equipment operation and maintenance cost, and create more benefits for power generation enterprises.
Traditional power plant equipment overhauls and has the condition of maintenance inadequately or maintaining excessively, and the realization of state monitoring method based on analytic model and knowledge needs a large amount of professional knowledge and field experience.
Disclosure of Invention
The invention aims to overcome the defects of insufficient maintenance or excessive maintenance and a state detection method in power plant equipment maintenance in the prior art, and provides an intelligent diagnosis and early warning method and system for a power plant.
In order to achieve the purpose, the invention adopts the following technical scheme to realize the purpose:
an intelligent diagnosis and early warning method for a power plant comprises the following steps:
step 1), acquiring real-time power plant SIS data, and preprocessing the acquired power plant SIS data;
step 2) constructing a data model by utilizing data in the power plant SIS historical database and the DCS database;
step 3) fault detection and fault diagnosis are carried out on the basis of the preprocessed real-time power plant SIS data and the data model;
and 4) constructing an evaluation system of the power plant operation system, calculating an evaluation benchmark and a failure threshold of the evaluation system, establishing an early warning model of the equipment, and calculating to obtain failure time of the equipment.
Preferably, in the step 1), a synchronous access mode and an asynchronous access mode are respectively adopted, the power plant SIS data operated on site are collected, and the collected power plant SIS data are stored in a real-time database.
Preferably, in the step 1), the power plant SIS data is preprocessed by using a clustering analysis method;
the preprocessing includes filtering and bad value culling.
Preferably, in step 2), different data modeling methods are adopted for different components and systems to construct data models;
and for the single-input single-output component, a data model is constructed by adopting a piecewise linearization method.
Preferably, the step of constructing the data model by using a piecewise linearization method specifically comprises the following steps:
where f is the output signal of the component, U is the input signal of the component, [ U [ [ U ]i,Ui+1]Denotes the i-th segment characteristic interval, k1、k2、k3The slopes of the input signal connecting lines at different stages in the model are respectively; b1、b2、b3Respectively constant of input signal connecting lines at different stages in the model;
the estimated values of the parameters of the model are represented,
is an output estimate of the model; j is the error deviation of the estimated output from the actual output.
Preferably, in step 3), the fault detection specifically includes:
assuming that the training set data under normal conditions is a matrix X ∈ Rn×mIt is decomposed as follows: x is WH + S,
wherein W ∈ Rn×kIs a basis matrix, H ∈ Rk×mWatch with clockShowing the matrix, S ∈ Rn×mIs a residual matrix; the optimization function is designed as follows:
s.t.W≥0,H≥0
the update rule of W, H, S by derivation and correlation theorem is as follows:
finally, the fault detection statistics are as follows:
S2＝diag(S(X)S(x))
wherein, S (X) represents a residual matrix obtained by using a training set, and S (x) represents a residual matrix obtained by using a testing set;
the fault judgment method comprises the following steps: calculating a control limit S according to a kernel density estimation method by using statistics obtained by calculation of a training setmThen, a statistical quantity is calculated using the test set, and if the statistical quantity exceeds a control limit, a fault is indicated.
An intelligent diagnosis and early warning system for power plant comprises
The data acquisition unit is used for acquiring real-time power plant operation data and preprocessing the acquired power plant operation data;
the model establishing unit is used for interacting with the data acquiring unit, dividing the components and the subsystems and establishing a data model;
the fault detection and diagnosis unit is interacted with the model building unit and carries out fault detection and fault diagnosis based on the output data of the data model and the real-time power plant SIS data;
and the early warning evaluation unit is interacted with the fault detection and diagnosis unit, constructs an early warning system, establishes an evaluation system of the early warning system, calculates an evaluation standard and a failure threshold of the evaluation system, establishes an early warning model, and calculates failure time.
Preferably, a visualization unit is further included for monitoring the operating state of the power plant.
Preferably, a control unit is included, the control unit comprising three control gates;
the control gates are a forgetting gate, an input gate and an output gate respectively;
the forgetting gate is used for screening and reserving important parts in long-term memory;
the input gate is used for selecting and updating the current short-term information;
and the output gate summarizes the information of the first two and outputs the information.
Compared with the prior art, the invention has the following beneficial effects:
the invention discloses an intelligent diagnosis method applied to a power plant. The data access module realizes reading operation with a power plant database; the diagnosis and early warning analysis module constructs a data model according to historical data, then diagnoses early fault germination by utilizing a robust nonnegative matrix factorization algorithm based on model output and real-time data, determines a failure threshold value by combining the operation performance of the system, and realizes prediction of the complete failure time of the system based on Gamma degradation modeling; the visual interaction module is used for displaying the current monitoring state of the system and the future fault trend. The invention realizes the fault detection of the system and equipment of the power plant based on the angle of data driving, utilizes abundant data resources stored in the power plant and based on the big data analysis technology, evaluates the performance of the system and equipment, predicts the future operation condition of the system or equipment based on the fault detection and the performance evaluation, and predicts the failure time of the system or equipment, thereby providing reasonable maintenance and repair suggestions for field personnel and greatly improving the working efficiency.
The invention also discloses an intelligent diagnosis and early warning system of the power plant, which is characterized in that a data-driven method is utilized, the state information of the evaluation system, equipment degradation and faults is obtained through mining analysis of an intelligent algorithm based on rich historical and real-time data in a power plant database, and the early warning can be carried out on the future performance decline. The data access module is designed to read the operation data of each device and system stored in the power plant database. The diagnosis and early warning analysis module completes the functions of modeling, fault detection and fault early warning according to the data, and is specifically represented as establishing a data model of a power plant system and equipment; detecting faults of the system and equipment by using the model and the real-time data; the method comprises the steps of establishing a performance evaluation system of the system and the equipment based on a data analysis method, determining evaluation benchmarks and failure thresholds of the system and the equipment in the system, finally establishing an early warning model, and predicting failure time of the system and the equipment according to the evaluation system. The visual interaction module is an interaction interface for displaying the system and equipment information of the power plant and the information of detection, evaluation, early warning and the like in front of a user.
Drawings
FIG. 1 is a flow chart of an intelligent diagnosis and early warning method for a power plant of the invention;
FIG. 2 is a functional implementation logic of the intelligent diagnosis and early warning method for the power plant of the invention;
FIG. 3 is a flow chart of piecewise linear modeling in an embodiment of the present invention;
FIG. 4 is a diagram of an LSTM basic unit of the intelligent diagnosis and early warning method for a power plant of the invention;
FIG. 5 is a visualized content of the intelligent diagnosis and early warning method for the power plant.
Detailed Description
The invention is described in further detail below with reference to the accompanying drawings:
example 1
An intelligent diagnosis and early warning method for a power plant comprises the following steps:
step 1), acquiring real-time power plant SIS data, and preprocessing the acquired power plant SIS data;
step 2) constructing a data model by utilizing data in the power plant SIS historical database and the DCS database;
step 3) fault detection and fault diagnosis are carried out on the basis of the preprocessed real-time power plant SIS data and the data model;
and 4) constructing an evaluation system of the power plant operation system, calculating an evaluation benchmark and a failure threshold of the evaluation system, establishing an early warning model of the equipment, and calculating to obtain failure time of the equipment.
Example 2
Taking a control loop of a certain thermal power generating unit as an example, as shown in fig. 1, the system structure diagram of the invention is composed of a data access module, a diagnosis and early warning analysis module and a visual interaction module. The system is characterized in that historical and real-time data stored in a power plant SIS database, a DCS database and the like are read, a data model of an object (component equipment and a subsystem) is established by a data driving method, and a big data analysis method is used for detecting the occurrence of faults according to model output and actual data; in addition, the performance degradation process of the object is predicted by formulating a performance evaluation reference, so that the early warning of the fault is realized; and finally, monitoring and operating field personnel by designing a visual client.
Specifically, in a data access module, on the basis of an OPC industrial standard, a field data acquisition technology is researched respectively for a synchronous access mode and an asynchronous access mode, an OPC client is further developed, and acquired field data are dumped to a real-time database; and secondly, based on a clustering analysis method, smoothing of random errors of the field data and detection of gross errors are realized, so that filtering and bad value elimination of the field data are realized.
The diagnosis and early warning analysis module mainly comprises the functions of modeling, detection, evaluation, early warning and the like.
First, different data modeling methods are adopted for different components and systems.
Aiming at a simple single-input single-output component, modeling is carried out by adopting a piecewise linearization method;
definition f denotes the output signal of the component, U is the input signal of the component, [ Ui, Ui+1]Representing the characteristic interval of the i-th section, and constructing the following components by using a least square methodSegment linear model:
assuming that the number of samples is N, the following model performance indexes are defined:
where f is the output signal of the component, U is the input signal of the component, [ U [ [ U ]i,Ui+1]Denotes the i-th segment characteristic interval, k1、k2、k3The slopes of the input signal connecting lines at different stages in the model are respectively; b is a mixture of1、b2、b3Respectively constant of input signal connecting lines at different stages in the model;
the estimated values of the parameters of the model are represented,
is an output estimate of the model; j is the error deviation of the estimated output from the actual output.
And when the performance index of the model is not higher than the set threshold value, the constructed model is approved, otherwise, the characteristic interval needs to be shortened, and the model is modeled again.
Aiming at a complex multiple input multiple output system, a long short term memory network (LSTM) is adopted to train a model; it contains three control gates: a forgetting gate, an input gate and an output gate; the forgetting gate is used for screening and reserving important parts in long-term memory; the input gate is used for selecting and updating the current short-term information; finally, the output door collects and outputs the information of the first two; when the input data is x, the memory cell is c, and the prediction output is h, first, the value of the candidate memory cell at the current time is calculated:
in the formula, Wxc、WhcThe weighted values are respectively input data and the weighted values output by the LSTM unit at the last moment; x is the number oftIs the input data value at time t, ht-1Predicting an output value for the time t-1;
the value of the input gate is then calculated:
it＝σ(Wxixt+Whiht-1+Wcict-1+bi)
in the formula, Wxi、Whi、WciRespectively the weight values of the input data, the prediction output data and the memory cell in the input gate, biIs offset, sigma is a common sigmoid function, and the value range is (0, 1);
calculating the value of the forgetting gate:
ft＝σ(Wxfxt+Whfht-1+Wcfct-1+bf)
in the formula, Wxf、Whf、WcfRespectively, the weight values of the input data, the predicted output data and the memory cell in the forgetting gate, bfIs an offset;
calculating the current memory cell state:
calculate the value of the output gate:
ot＝σ(Wxoxt+Whoht-1+Wcoct-1+bo)
in the formula, Wxo、Who、WcoRespectively the weight values of the input data, the predicted output data and the memory cell in the output gate, boIs an offset;
finally, the output of the LSTM is calculated:
ht＝ot tanh Ct
after the piecewise linear model and the LSTM model are trained, the control signal of the system or the component is input to obtain the standard output information of the object.
Secondly, the method for realizing the fault detection function is as follows:
assuming that the training set data (i.e. the real-time signal as the model input and the obtained standard object output signal) under normal conditions is a matrix X ∈ Rn×mIt is decomposed as follows:
X＝WH+S
wherein W ∈ Rn×kIs a basis matrix, H ∈ Rk×mIs a representation matrix, S ∈ Rn×mIs a residual matrix; the optimization function is designed as follows:
s.t.W≥0,H≥0
the update rule of W, H, S by derivation and correlation theorem is as follows:
finally, the fault detection statistics are as follows:
S2＝diag(S(X)S(x))
wherein, S (X) represents a residual matrix obtained by using a training set, and S (x) represents a residual matrix obtained by using a testing set; the fault judgment method comprises the following steps: calculated using a training setThe statistic calculates the control limit S according to the kernel density estimation methodmThen, a statistical quantity is calculated using the test set, and if the statistical quantity exceeds a control limit, a fault is indicated.
Next, Hurst index was used to evaluate the performance of the system:
for a time series Y containing N data, the average value thereof is calculated
Calculating the sequence of dispersion accumulation of Y'
Dividing the dispersion accumulation sequence Y' into W non-overlapping equal-length intervals by a window length L, (W is N/L and takes an integer); for each interval, a first order linear fit is performed on the L data points contained in the interval by using a least square method:
if L data points are (t (1), y (1)), (t (2), y (2)), …, (t (k), y (k)), …, (t (L), y (L)) in this order, there are
The fitting result on the jth interval is yj＝ajt+bj(j＝1,2,3,...,W)；
Calculating the sum of the mean square deviations after the filtering trend of the jth interval:
calculating the DFA fluctuation function F (L):
taking different window lengths L to obtain multiple groups (L, F (L)), F (L) and L satisfy the power law relation:
F(L)＝a×Lα
in the log-log coordinates (ln (L), ln (F (L)), the data points are fitted by the least squares method, there
lnF(L)＝αln(L)+lna
Wherein the slope α of the straight line portion, i.e., Hurst index;
if alpha is 0.5, the system performance is better; whereas the more it deviates from 0.5, the worse the performance; therefore, based on a large amount of historical data, the Hurst indexes under different conditions are calculated in a statistical mode, the optimal evaluation reference is screened out, and the index threshold value l of system performance failure is determined.
Finally, the method for early warning system failure comprises the following steps:
the system degradation model is established as follows:
Dm(t)＝D(t)+ε(t)
wherein D ism(t) is the Hurst performance index measured value of the system at the measuring time t, D (t) is the Hurst performance index actual value of the system at the measuring time t,
d (t) obeys a Gamma distribution with a probability density function of:
wherein,
the estimated values of the parameters were calculated as follows:
α＝E(Δω(t))/β
σ2＝0.5(E(Δω(t)2)-E(Δω(t))2-βE(Δω(t)))
then, the predicted system life distribution is as follows:
and obtaining the fault or failure time of the system or the component under the specified failure threshold value according to the formula.
The visual interaction module displays diagnosis and early warning information of the power plant through a mode issued by the B/S or the C/S, wherein the diagnosis and early warning information includes but is not limited to modeling results, real-time state monitoring, performance evaluation, fault early warning and the like of components and systems. And the analysis and maintenance of field operators are facilitated by presenting in the forms of graph curves and the like.
In FIG. 4, the input data is x, the memory cell is c, the prediction output is h, the first σ is the input gate, the second σ is the forgetting gate, the third σ is the output gate, "+" indicates the current memory cell, and the left tanh indicates the candidate memory cell
First a data stream xt,ht-1,ct-1Via the input gate, the following are obtained:
it＝σ(Wxixt+Whiht-1+Wcict-1+bi)
through forgetting the door and obtaining:
ft＝σ(Wxfxt+Whfht-1+Wcfct-1+bf)
finally, input gate output, ct-1And the forgetting gate output and the candidate unit are subjected to multiplication calculation, and are converged after the plus to obtain the state of the current memory unit:
data stream xt,ht-1,ct-1Through the output gate, the following are obtained:
ot＝σ(Wxoxt+Whoht-1+Wcoct-1+bo)
and finally, outputting the LSTM output after X calculation by the current memory unit and the output gate:
ht＝ot tanh Ct
example 3
An intelligent diagnosis and early warning system for power plant comprises
The data acquisition unit is used for acquiring real-time power plant operation data and preprocessing the acquired power plant operation data;
the model establishing unit is used for interacting with the data acquiring unit, dividing the components and the subsystems and establishing a data model;
the fault detection and diagnosis unit is interacted with the model building unit and carries out fault detection and fault diagnosis based on the output data of the data model and the real-time power plant SIS data;
and the early warning evaluation unit is interacted with the fault detection and diagnosis unit, constructs an early warning system, establishes an evaluation system of the early warning system, calculates an evaluation standard and a failure threshold of the evaluation system, establishes an early warning model, and calculates failure time.
And the visualization unit is used for monitoring the operating state of the power plant.
The control unit comprises three control doors;
the control gates are a forgetting gate, an input gate and an output gate respectively; the forgetting gate is used for screening and reserving important parts in long-term memory; the input gate is used for selecting and updating the current short-term information; and the output gate summarizes the information of the first two and outputs the information.
In conclusion, the fault detection method based on the big data analysis technology realizes fault detection of the system and the equipment of the power plant based on the angle of data driving, utilizes abundant data resources stored in the power plant, and based on the big data analysis technology, estimates the performance of the system and the equipment, predicts the future operation condition of the system or the equipment based on the fault detection and the performance evaluation, and predicts the failure time of the system or the equipment with the fault, so that reasonable maintenance and repair suggestions are provided for field personnel, and the working efficiency is greatly improved.
The above contents are only for illustrating the technical idea of the present invention, and the protection scope of the present invention should not be limited thereby, and any modification made on the basis of the technical idea proposed by the present invention falls within the protection scope of the claims of the present invention.
Claims (9)
1. The intelligent diagnosis and early warning method for the power plant is characterized by comprising the following steps:
step 1), acquiring real-time power plant SIS data, and preprocessing the acquired power plant SIS data;
step 2) constructing a data model by utilizing data in the power plant SIS historical database and the DCS database;
step 3) fault detection and fault diagnosis are carried out on the basis of the preprocessed real-time power plant SIS data and the data model;
and 4) constructing an evaluation system of the power plant operation system, calculating an evaluation benchmark and a failure threshold of the evaluation system, establishing an early warning model of the equipment, and calculating to obtain failure time of the equipment.
2. A power plant intelligent diagnosis and early warning method according to claim 1, characterized in that in step 1), a synchronous access mode and an asynchronous access mode are respectively adopted, power plant SIS data operated on site are collected, and the collected power plant SIS data are stored in a real-time database.
3. A power plant intelligent diagnosis and early warning method according to claim 1, characterized in that in the step 1), a clustering analysis method is used for preprocessing the power plant SIS data;
the preprocessing includes filtering and bad value culling.
4. A power plant intelligent diagnosis and early warning method according to claim 1, characterized in that in the step 2), different data modeling methods are adopted for different components and systems to construct data models;
and for the single-input single-output component, a data model is constructed by adopting a piecewise linearization method.
5. A power plant intelligent diagnosis and early warning method according to claim 4, characterized in that the construction of the data model by using a piecewise linearization method specifically comprises:
where f is the output signal of the component, U is the input signal of the component, [ U [ [ U ]i,Ui+1]Denotes the i-th segment characteristic interval, k1、k2、k3The slopes of the input signal connecting lines at different stages in the model are respectively; b1、b2、b3Respectively constant of input signal connecting lines at different stages in the model;
representing the estimated values of the parameters of the model,
is an output estimate of the model; j is the error deviation of the estimated output from the actual output.
6. A power plant intelligent diagnosis and early warning method according to claim 5, wherein in the step 3), the fault detection specifically comprises:
assuming that the training set data under normal conditions is a matrix X ∈ Rn×mIt is decomposed as follows: x is WH + S,
wherein W ∈ Rn×kIs a basis matrix, H ∈ Rk×mIs a representation matrix, S ∈ Rn×mIs a residual matrix; the optimization function is designed as follows:
s.t.W≥0,H≥0
the update rule of W, H, S by derivation and correlation theorem is as follows:
finally, the fault detection statistics are as follows:
S2＝diag(S(X)S(x))
wherein, S (X) represents a residual matrix obtained by using a training set, and S (x) represents a residual matrix obtained by using a testing set;
the fault judgment method comprises: calculating a control limit S according to a kernel density estimation method by using statistics obtained by calculation of a training setmThen, a statistical quantity is calculated using the test set, and if the statistical quantity exceeds a control limit, a fault is indicated.
7. An intelligent diagnosis and early warning system for a power plant is characterized by comprising
The data acquisition unit is used for acquiring real-time power plant operation data and preprocessing the acquired power plant operation data;
the model establishing unit is used for interacting with the data acquiring unit, dividing the components and the subsystems and establishing a data model;
the fault detection and diagnosis unit is interacted with the model building unit and carries out fault detection and fault diagnosis based on the output data of the data model and the real-time power plant SIS data;
and the early warning evaluation unit is interacted with the fault detection and diagnosis unit, constructs an early warning system, establishes an evaluation system of the early warning system, calculates an evaluation standard and a failure threshold of the evaluation system, establishes an early warning model, and calculates failure time.
8. A power plant intelligent diagnosis and early warning system according to claim 7, further comprising a visualization unit for monitoring the operating status of the power plant.
9. The power plant intelligent diagnosis and early warning system of claim 7, comprising a control unit, wherein the control unit comprises three control doors;
the control gates are a forgetting gate, an input gate and an output gate respectively;
the forgetting gate is used for screening and reserving important parts in long-term memory;
the input gate is used for selecting and updating the current short-term information;
and the output gate summarizes the information of the first two and outputs the information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111408475.0A CN114519923B (en) | 2021-11-24 | Intelligent diagnosis and early warning method and system for power plant |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111408475.0A CN114519923B (en) | 2021-11-24 | Intelligent diagnosis and early warning method and system for power plant |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114519923A true CN114519923A (en) | 2022-05-20 |
| CN114519923B CN114519923B (en) | 2024-09-24 |
Family
ID=
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115081926A (en) * | 2022-07-14 | 2022-09-20 | 石家庄良村热电有限公司 | Operation safety early warning method and system suitable for intelligent power plant |
| CN115639470A (en) * | 2022-09-23 | 2023-01-24 | 贵州北盘江电力股份有限公司光照分公司 | Generator monitoring method and system based on data trend analysis |
| CN116340427A (en) * | 2023-04-25 | 2023-06-27 | 北京工业大学 | Early warning method and system for environment-friendly data |
| CN117519067A (en) * | 2023-10-20 | 2024-02-06 | 东北大学 | Multi-frame control performance evaluation method in continuous rolling process |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108460207A (en) * | 2018-02-28 | 2018-08-28 | 上海华电电力发展有限公司 | A kind of fault early warning method of the generating set based on operation data model |
| CN109657836A (en) * | 2018-11-19 | 2019-04-19 | 东莞理工学院 | A kind of Power System Analysis method for early warning based on big data |
| CN112257988A (en) * | 2020-09-29 | 2021-01-22 | 中广核工程有限公司 | Complex accident feature identification and risk early warning system and method for nuclear power plant |
| CN113065580A (en) * | 2021-03-17 | 2021-07-02 | 国能大渡河大数据服务有限公司 | Power plant equipment management method and system based on multi-information fusion |
| CN113189941A (en) * | 2021-03-16 | 2021-07-30 | 珠海市钰海电力有限公司 | Intelligent fault early warning method and system for power plant power generation equipment |
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108460207A (en) * | 2018-02-28 | 2018-08-28 | 上海华电电力发展有限公司 | A kind of fault early warning method of the generating set based on operation data model |
| CN109657836A (en) * | 2018-11-19 | 2019-04-19 | 东莞理工学院 | A kind of Power System Analysis method for early warning based on big data |
| CN112257988A (en) * | 2020-09-29 | 2021-01-22 | 中广核工程有限公司 | Complex accident feature identification and risk early warning system and method for nuclear power plant |
| CN113189941A (en) * | 2021-03-16 | 2021-07-30 | 珠海市钰海电力有限公司 | Intelligent fault early warning method and system for power plant power generation equipment |
| CN113065580A (en) * | 2021-03-17 | 2021-07-02 | 国能大渡河大数据服务有限公司 | Power plant equipment management method and system based on multi-information fusion |
Non-Patent Citations (1)
| Title |
|---|
| 邱凤翔;徐治皋;司风琪;赵跃;: "基于主元分析的关联规则挖掘及在电厂中的应用", 电力系统自动化, no. 16, 25 August 2009 (2009-08-25) * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115081926A (en) * | 2022-07-14 | 2022-09-20 | 石家庄良村热电有限公司 | Operation safety early warning method and system suitable for intelligent power plant |
| CN115639470A (en) * | 2022-09-23 | 2023-01-24 | 贵州北盘江电力股份有限公司光照分公司 | Generator monitoring method and system based on data trend analysis |
| CN115639470B (en) * | 2022-09-23 | 2024-01-30 | 贵州北盘江电力股份有限公司光照分公司 | Generator monitoring method and system based on data trend analysis |
| CN116340427A (en) * | 2023-04-25 | 2023-06-27 | 北京工业大学 | Early warning method and system for environment-friendly data |
| CN116340427B (en) * | 2023-04-25 | 2023-10-13 | 北京工业大学 | Method for environmental protection data early warning system |
| CN117519067A (en) * | 2023-10-20 | 2024-02-06 | 东北大学 | Multi-frame control performance evaluation method in continuous rolling process |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111275288A (en) | XGboost-based multi-dimensional data anomaly detection method and device | |
| CN113036913B (en) | Method and device for monitoring state of comprehensive energy equipment | |
| CN117234785B (en) | Centralized control platform error analysis system based on artificial intelligence self-query | |
| CN112101431A (en) | Electronic equipment fault diagnosis system | |
| CN113339204B (en) | Wind driven generator fault identification method based on hybrid neural network | |
| CN117290800B (en) | Timing sequence anomaly detection method and system based on hypergraph attention network | |
| CN109492790A (en) | Wind turbines health control method based on neural network and data mining | |
| CN116050665B (en) | Heat supply equipment fault prediction method | |
| CN115438726A (en) | Device life and fault type prediction method and system based on digital twin technology | |
| CN117142009B (en) | Scraper conveyor health state assessment method based on graph rolling network | |
| CN118157132B (en) | Data mining method and device for voltage monitoring system based on neural network | |
| CN117408162B (en) | Power grid fault control method based on digital twin | |
| CN112906288A (en) | Switch cabinet state evaluation method based on normal sample data | |
| CN114925718A (en) | Transformer core grounding current online prediction method and device, storage medium and fault diagnosis method | |
| CN115081647A (en) | Industrial intelligent instrument fault pre-diagnosis method based on Bayesian network model | |
| CN116821610B (en) | Method for optimizing wind power generation efficiency by utilizing big data | |
| CN118035814A (en) | State fault early warning method and system based on digital twin field data analysis equipment | |
| CN117493798A (en) | Meteorological environment data analysis method and system | |
| CN110455370B (en) | Flood-prevention drought-resisting remote measuring display system | |
| CN114519923B (en) | Intelligent diagnosis and early warning method and system for power plant | |
| CN114519923A (en) | Intelligent diagnosis and early warning method and system for power plant | |
| CN115600695A (en) | Fault diagnosis method of metering equipment | |
| CN114332515A (en) | Online evaluation method for operation safety of coal mill based on intuition fuzzy clustering | |
| CN117804811B (en) | Health diagnosis system and method for deep neural network port machinery equipment of attention mechanism | |
| CN118096131B (en) | Operation and maintenance inspection method based on electric power scene model |
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 |