CN116295948B - Abnormality detection method, system and storage medium of industrial temperature sensor in large temperature difference environment - Google Patents
Abnormality detection method, system and storage medium of industrial temperature sensor in large temperature difference environment Download PDFInfo
- Publication number
- CN116295948B CN116295948B CN202310261906.8A CN202310261906A CN116295948B CN 116295948 B CN116295948 B CN 116295948B CN 202310261906 A CN202310261906 A CN 202310261906A CN 116295948 B CN116295948 B CN 116295948B
- Authority
- CN
- China
- Prior art keywords
- temperature
- temperature sensor
- industrial
- detected
- module
- 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.)
- Active
Links
- 230000005856 abnormality Effects 0.000 title claims abstract description 39
- 238000001514 detection method Methods 0.000 title claims abstract description 23
- 238000003860 storage Methods 0.000 title claims abstract description 9
- 230000002159 abnormal effect Effects 0.000 claims abstract description 48
- 238000013507 mapping Methods 0.000 claims abstract description 38
- 238000000034 method Methods 0.000 claims description 37
- 238000012360 testing method Methods 0.000 claims description 24
- 238000004458 analytical method Methods 0.000 claims description 12
- 230000001364 causal effect Effects 0.000 claims description 6
- 238000010998 test method Methods 0.000 claims description 5
- 238000002347 injection Methods 0.000 claims description 3
- 239000007924 injection Substances 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 2
- 238000012512 characterization method Methods 0.000 description 13
- 238000010586 diagram Methods 0.000 description 8
- 230000008901 benefit Effects 0.000 description 7
- 238000004590 computer program Methods 0.000 description 7
- 238000012216 screening Methods 0.000 description 7
- 230000008859 change Effects 0.000 description 6
- 238000013461 design Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000012545 processing Methods 0.000 description 6
- 238000004519 manufacturing process Methods 0.000 description 5
- 238000007781 pre-processing Methods 0.000 description 5
- 229910000831 Steel Inorganic materials 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000010801 machine learning Methods 0.000 description 4
- 239000010959 steel Substances 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 230000036541 health Effects 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 230000009194 climbing Effects 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000011031 large-scale manufacturing process Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000000528 statistical test Methods 0.000 description 1
- 238000009628 steelmaking Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
- G01K15/00—Testing or calibrating of thermometers
- G01K15/007—Testing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/04—Constraint-based CAD
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/08—Probabilistic or stochastic CAD
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/08—Thermal analysis or thermal optimisation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Physics (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Algebra (AREA)
- Probability & Statistics with Applications (AREA)
- Operations Research (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
The invention relates to the technical field of sensor abnormality detection, in particular to an abnormality detection method, an abnormality detection system and a storage medium of an industrial temperature sensor in a large-temperature-difference environment. When the operation parameters to be detected are abnormal, the operation parameters to be detected in the high-temperature, low-temperature and variable-temperature states can be comprehensively decided through the mapping connection relation network, and the abnormal probability of each module component of the industrial temperature sensor is judged. The invention can greatly reduce the expenditure of computing resources, reduce the operation load of the server, and further improve the operation stability of the server. According to the invention, the judgment is carried out according to the operation parameters in the high-temperature and low-temperature variable-temperature states, and compared with a model constructed in a temperature state, the diversity of data samples is increased, and the constructed mapping connection relation network is more complete and accurate.
Description
Technical Field
The invention relates to the technical field of sensor abnormality detection, in particular to an abnormality detection method, an abnormality detection system and a storage medium of an industrial temperature sensor in a large-temperature-difference environment.
Background
After the industrial temperature sensor is developed in a long-term way, the industrial temperature sensor has higher intelligence and operation reliability, and can be normally and stably used for a specified service life under the condition that the temperature condition of the working environment is stable. However, for the industrial temperature sensor which works in the extreme environment temperature condition, the abnormal or fault phenomenon is still endangered, for example, when the industrial temperature sensor works in the steelmaking link, the environment temperature of the industrial temperature sensor can reach 2000 ℃ at the highest time in the daytime when the blast furnace ironmaking process is carried out, and the environment temperature of the industrial temperature sensor gradually returns to the daily environment temperature when the process is stopped at night, and the severe temperature change can damage the parts of the industrial temperature sensor, so that the abnormal or fault of the industrial temperature sensor is gradually caused. Meanwhile, the equipment complexity of the industrial temperature sensor is greatly improved compared with the prior art, the industrial temperature sensor is not only composed of a simple temperature measuring device, but also an intelligent industrial temperature sensor/instrument integrating various components such as an intelligent control module, a data storage module, a digital display module, a temperature measuring module and the like, the component modules together form a system, each component module is an independent subsystem and is composed of various parts, and the abnormal or damaged part of each part can directly or indirectly influence the normal operation of the subsystem in which the component module is positioned, so that the influence on the whole system is caused. Thus, modeling analysis methods for conventional industrial temperature sensors are clearly no longer applicable to the intelligent industrial temperature sensors at hand, and a completely new method capable of coping with such a complex system is needed.
For an industrial temperature sensor working under the condition of severe environmental temperature change, due to the improvement of the current production and manufacturing process, the extreme temperature difference does not cause sudden abnormality or failure of the industrial temperature sensor, but damage caused by long-term working under such environment still cannot be avoided, and the damage can be gradually accumulated along with working time, so that the industrial temperature sensor is not easy to generate abrupt type abnormality, but is usually a gradual type abnormality mainly comprising trend abnormality. The trend abnormality does not have obvious influence on the performance and the like of the industrial temperature sensor in the initial stage, is more concealed, usually shows obvious abnormal operation state after long-time accumulation, is difficult to repair in time, further evolves into faults, influences the normal operation of production, forms potential safety hazards and the like. Therefore, how to discover the trend abnormality existing in the operation process of the industrial temperature sensor as early as possible is an important ring for optimizing and perfecting the health management system of the in-service industrial temperature sensor. For the traditional industrial temperature sensor, the structure is simple, the modeling analysis is simple, and the one-to-one physical reduction modeling of a 'entity-model' can be realized generally, so that a good analysis effect can be achieved; for an intelligent industrial temperature sensor, because of the complexity of system composition, physical modeling analysis becomes extremely difficult, and secret connection relations generally exist among all components and all subsystems, so that complete expression is difficult to obtain, the built system is in chaos caused by forced physical modeling, and the analysis effect is also generally poor.
Currently, modeling analysis methods for intelligent industrial temperature sensors are mainly data-driven methods, and among many data-driven methods, machine learning and neural networks are the most commonly used and effective methods. However, in the stage of anomaly detection, modeling methods such as machine learning and neural networks are too heavy, and when the number of industrial temperature sensors is large and the anomaly detection task is frequent, excessive computing resources are often occupied, and the requirement of rapid and quick anomaly detection cannot be met. In addition, for trend anomalies of interest in industrial temperature sensors, the monitoring time scale is generally longer, the data volume is larger, if methods such as machine learning and neural network are adopted, the required model training data volume may be extremely huge, and industrial temperature sensors in early or middle service stages often cannot provide trend anomaly data capable of meeting the requirement volume, so that the trend anomaly detection effect is affected.
Disclosure of Invention
The first object of the present invention is to provide an abnormality detection method for an industrial temperature sensor in a large temperature difference environment, which can detect the abnormality trend of the industrial temperature sensor and can determine the specific module of the industrial sensor in which the abnormality occurs.
The first object of the invention is achieved by the technical proposal that the method comprises the following steps:
collecting operation parameters to be detected of an industrial temperature sensor to be detected in a high-temperature and low-temperature variable-temperature state;
If the operation parameters to be detected are abnormal, comprehensively deciding the operation parameters to be detected in the high-temperature, low-temperature and variable-temperature states through a mapping connection relation network, and judging the abnormal probability of each module component of the industrial temperature sensor;
And the mapping connection relation network is used for mapping the fault probability relation between each operation parameter of the industrial temperature sensor in the high-temperature, low-temperature and variable-temperature states and the corresponding module component.
The design has the advantages that the most basic statistical analysis method in regression data driving is adopted, and the statistical characteristics of trend anomalies are pertinently improved and perfected, so that compared with modeling methods of machine learning, neural networks and the like, the cost of computing resources can be greatly reduced, the operation load of a server side is reduced, and the operation stability of the server side is further improved. The invention adopts the mapping connection relation network, not only can detect the abnormal trend of the industrial temperature sensor, but also can judge the specific module with the abnormality in the industrial sensor. According to the invention, the judgment is carried out according to the operation parameters in the high-temperature and low-temperature variable-temperature states, and compared with a model constructed in a temperature state, the diversity of data samples is increased, and the constructed mapping connection relation network is more complete and accurate. If the operation parameters in a single temperature state are adopted for judgment, the condition that one parameter is abnormal and corresponds to a plurality of modules is easy to occur, and further accurate judgment is not possible.
Further, the method for judging the operation abnormality of the operation parameters to be detected comprises the following steps:
recording normal operation parameters of the industrial temperature sensor in a normal working state under high temperature and low temperature and variable temperature states, and calculating normal operation mathematical expectations of the parameters;
and carrying out hypothesis testing analysis on the operation parameters to be detected and the normal operation mathematical expectation in sequence, and judging whether the industrial temperature sensor is abnormal or not.
The design has the advantages that whether the abnormality occurs or not is judged in a statistical mode, the calculated amount is smaller, the speed is faster, and the design is suitable for a large industrial sensor network.
Further, the method for establishing the mapping connection relation network comprises the following steps:
The repeated abnormal injection test is carried out on each module component of the industrial temperature sensor under the conditions of high temperature, low temperature and variable temperature respectively;
And respectively counting the abnormal occurrence times and the abnormal occurrence degree of each abnormal operation parameter with a mapping relation under the conditions of high temperature and low temperature and changing temperature, establishing the mapping relation between each operation parameter and each module component of the industrial temperature sensor by using a Granger causal relation test method according to the ratio of the statistical data, and determining the abnormal weight coefficient of each abnormal operation parameter and the corresponding module component under the condition of different temperatures.
The design has the advantages that the mapping connection relation network established by the Granger causality test method is more accurate in judging module faults under the conditions of high temperature and low temperature and variable temperature, and the Granger causality test method is small in calculated quantity and high in accuracy.
Further, the method for judging the abnormal probability of each module component of the industrial temperature sensor comprises the following specific steps:
when the industrial temperature sensor is in any single temperature state, the operation parameter to be detected passes through a mapping connection relation network, and the probability of failure of a specific module part is greater than a first threshold value, and the module part is directly judged to be failed;
When the industrial temperature sensor is in any single temperature state, the operation parameters to be detected are connected with the relation network through mapping, the probability of failure of the specific module part is smaller than a first threshold value, and when the probability of failure of the specific module part is larger than a second threshold value, the comprehensive judgment is carried out by combining the failure probabilities in the other two temperature states;
when the industrial temperature sensor is in any single temperature state, the operation parameter to be detected passes through the mapping connection relation network, and when the probability of failure of the specific module component is smaller than a second threshold value, the operation parameter is judged to be normal.
The design has the advantages that the three fault judgment conditions are adopted by combining three mapping relation networks under high temperature and low temperature and variable temperature, so that the judgment logic is tighter and the accuracy is high.
Further, the specific method for comprehensively judging by combining the fault probabilities in the other two temperature states is as follows:
And calculating the average probability of failure of the specific module component under three temperature states, and judging that the module component fails if the average probability is larger than a preset value.
The design has the advantages of simple judgment method and high calculation speed.
Further, the industrial temperature sensor comprises a temperature sensing module, a temperature transmitting module, a digital display module, a wireless transmitting module and a main control module.
Further, the operation parameters include:
resistance value, current value and voltage value when the temperature sensing module operates;
The temperature transmitting module is used for transmitting an input signal value and an output signal value when in operation;
the working voltage and the battery power of the digital display module during operation;
signal intensity, capacitance value and inductance value when the wireless transmitting module operates;
motherboard temperature and working voltage when the main control module operates.
Further, the specific method for carrying out hypothesis testing analysis on the operation parameters to be detected and the mathematical expectation of normal operation in sequence and judging whether the industrial temperature sensor is abnormal comprises the following steps:
the parameters are subjected to time sequence difference operation, difference normalization summation, difference sum variance operation, and test statistics and double-tail test.
The second object of the invention is to provide an abnormality detection system for an industrial temperature sensor in a large temperature difference environment.
The second object of the present invention is achieved by a scheme comprising:
the acquisition unit is used for acquiring operation parameters to be detected of the industrial temperature sensor to be detected in a high-temperature, low-temperature and variable-temperature state;
the abnormality judging unit is used for judging whether the operation parameters to be detected are abnormal;
Mapping connection relation network unit, if the operation parameter to be detected is abnormal, making comprehensive decision on the operation parameter to be detected in high temperature, low temperature and variable temperature state, judging the probability of abnormality of each module component of the industrial temperature sensor
It is a third object of the present invention to provide a storage medium storing a number of instructions adapted to be loaded by a processor to perform the above-mentioned anomaly detection method.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof.
Drawings
The drawings of the present invention are described below.
FIG. 1 is a schematic flow chart of the present invention;
FIG. 2 is a schematic diagram of a process for screening and preprocessing operating parameters;
fig. 3 is a schematic diagram of a mapping connection relationship network between each parameter of the target temperature sensor with an abnormal weight coefficient and each component of each module.
Detailed Description
The invention is further described below with reference to the drawings and examples.
In the embodiment, a wireless temperature sensor in an SBW integrated series produced by Chongqing Sichuan instrument automation Co., ltd is selected. The temperature sensor has higher intelligent degree, more complex system composition, and consists of a temperature sensing module, a temperature transmitting module, a digital display module, a wireless transmitting module and a main control module, and is widely used in a plurality of large-scale production scenes such as Beijing Tang steel, bao steel, saddle steel, climbing steel and the like, and the temperature change of the production links is larger. The temperature sensor can be divided according to a system level, namely a sensor system layer and a sensor module layer, wherein the sensor system layer is an SBW integrated wireless temperature sensor complete machine system, and the sensor module layer is sequentially an SBW integrated wireless temperature sensor temperature sensing module, a temperature transmitting module, a digital display module, a wireless transmitting module and a main control module.
As shown in fig. 1, the abnormality detection procedure for the SBW integrated wireless temperature sensor is as follows:
And step 1, acquiring normal operation parameters of the SBW integrated wireless temperature sensor in a normal working state under high temperature, low temperature and variable temperature states.
The normal operation parameters comprise all parameter data which can be read in the SBW integrated wireless temperature sensor system, and of course, the normal operation parameters can also be other parameter data which are specially designated by an expert. The normal operation parameters are parameter operation data sets in the initial service period and the period to be tested of the target temperature sensor, and the parameter operation data sets comprise data values of all parameters related to the target temperature sensor, which can be monitored by operation and maintenance personnel. The initial stage of service refers to a period of time calculated from when the target temperature sensor is put into use and starts to work normally, the target temperature sensor in the initial stage of service is at a normal health level, and no abnormal running trend exists.
And 2, screening and preprocessing the obtained normal operation parameters, as shown in fig. 2, and respectively solving the mathematical expectations of normal operation under the high-temperature, low-temperature and variable-temperature states.
The screening is to remove parameter data which is irrelevant to the SBW integrated wireless temperature sensor system from the obtained normal operation parameter data, such as temperature data, temperature range, working time and the like of a target monitored by the SBW integrated wireless temperature sensor; the operation parameters related to the SBW integrated wireless temperature sensor are reserved, such as a resistance value, a current value and a voltage value of a temperature sensing module, an input signal value and an output signal value of a temperature transmitting module, a working voltage and a battery electric quantity of a digital display module, a signal intensity and a capacitance value of the wireless transmitting module, an inductance value of the wireless transmitting module, a main board temperature and a working voltage of a main control module and the like.
The preprocessing is mainly to perform HART protocol analysis and CRC check on the normal operation parameters of the screened SBW integrated wireless temperature sensor, convert the acquired binary data into decimal data which can be directly analyzed and utilized, then reject significant nonstandard data such as noise interference and the like in the data, and finally sort the acquired normal operation parameters according to time.
The screening and preprocessing are conventional processing based on the acquired data, and are not necessary, and conventional replacement processing can be performed according to the special data.
The normal operation mathematical expectation under the high-temperature and low-temperature and variable-temperature states is used as a reference value of each operation parameter of the SBW integrated wireless temperature sensor under the high-temperature and low-temperature and variable-temperature states; of course, other data of the SBW integrated wireless temperature sensor in the high temperature, low temperature and variable temperature states respectively can be used as reference values.
And step 3, collecting to-be-detected operation parameters of the SBW integrated wireless temperature sensor in high temperature, low temperature and variable temperature states, and executing screening and preprocessing the to-be-detected operation parameters, which are the same as those in the step 2.
The time length of the period to be measured is determined by operation staff according to the abnormal detection requirement of the operation trend of the target temperature sensor in the actual operation process and the actual acquisition frequency of each parameter data of the target temperature sensor. The data acquisition frequency of each operation parameter of the SBW integrated wireless temperature sensor is set to be 5 seconds once, and the time period length is 1 hour, so that each acquired parameter data set contains 720 parameter data points in total.
And 4, carrying out anomaly detection on each operation parameter to be detected.
Carrying out difference value operation on all data points and data corresponding to the subsequent time points, combining the normal operation mathematical expectation of each parameter, normalizing the data points to be-1, 0 and 1 according to different ranges of the difference values, and carrying out summation operation on all the normalized difference values; determining the number of repeated data groups in the data sequence and the number of repeated data in each group, and calculating the variance of the difference sum according to the number of repeated data groups; and combining the normal operation mathematical expectation of each parameter, carrying out different mathematical transformations on the difference value sum and the variance thereof according to the difference value sum and the difference range to obtain test statistics, and carrying out double-tail test on the test statistics to obtain the trend abnormal condition of a single parameter. So far, the hypothesis testing analysis of the single parameter operation data is completed. The normal operation mathematical expected setting range of each parameter refers to dividing the real number domain into three parts, namely negative infinity to 2% of opposite number of each parameter mathematical expected absolute value, 2% of opposite number of each parameter mathematical expected absolute value to 2% of each parameter mathematical expected absolute value and 2% of each parameter mathematical expected absolute value to positive infinity.
And (3) setting a single parameter as p, and executing the operation in the step (2) on the single parameter, wherein the mathematical expectation of normal operation of the single parameter under three temperature conditions of the initial service period of the SBW integrated wireless temperature sensor is calculated to be E h、En、Em. And (3) marking the operation parameters to be detected obtained in the step (3) as X, X= (X 1,x2,…,x720), marking the corresponding time sequence as T, T= (T 1,T2,…,T720), wherein the time points in T are arranged according to time ascending sequence, T epsilon T h∪Tn∪Tm, T h represents the high-temperature working period of the SBW integrated wireless temperature sensor, T n represents the normal-temperature working period of the SBW integrated wireless temperature sensor, and T m represents the variable-temperature working period of the SBW integrated wireless temperature sensor.
Step 4.1, performing a difference operation on data (X n+1,xn+2,…,x720) corresponding to all time points after each data point X n in the operation parameter set X to be detected, namely (X n+1-xn,xn+2-xn,…,x720-xn), wherein n is less than or equal to 719. Finally get together and containI.e. 258840 difference points, denoted D, d= (D 1,d2,…,d258840).
Step 4.2, normalizing each difference point D n in the set D, and normalizing D n to 1 if D n > 0.02|e|; if d n < -0.02|E|, then normalize d n to-1; if d n > -0.02|E|andd n <0.02|E|, then d n is normalized to 0. Wherein the value of E depends on the period of time in which the time sequence T is located, if T E T h, e=e h; if T E T n, e=e n; if T E T m, then e=e m;Em＝(Eh+En)/2. Finally, a normalized difference point set is obtained and is marked as D ', D' = (D '1,d'2,…,d'258840), wherein the value range of D' n is { -1,0,1}. All the difference points in the set of difference points D 'are summed and the result is noted S, s=d' 1+d'2+…+d'258840.
Step 4.3, determining the number of repeated data groups in the operation parameter set X to be detected, which is marked as M, and the number of repeated data in each group, which is marked as N= (N 1,n2,…,nk). For example, if there is x1＝x2＝x3＝C1,x6＝x13＝x29＝x55＝C2,x77＝x82＝C3, in the parameter data set X where C 1,C2,C3 is a constant, the number of sets of data that repeatedly occur in the parameter data set X, m=3, and the number of times data within each set is repeated, n= (N 1,n2,n3), where N 1＝3,n2＝4,n3 =2.
Step 4.4, calculating the variance of the difference and S on the basis of step 4.2 and step 4.3, denoted VAR (S), For this embodiment, namely The variance VAR (S) is transformed to obtain a test statistic Z, if S >0.02|E|If S < -0.02|E|, thenIf S > -0.02|e| and S <0.02|e|, then z=0. Wherein the value method of E is the same as that of the step 4.2. In the case where the parameter data set X sample size is n=720, the test statistic Z approximately follows a standard gaussian distribution.
And 4.5, performing double-tail test on the test statistic Z. The assumption is that: under the significance level alpha, the operation trend abnormality exists in the to-be-detected operation parameter set X acquired by the current parameter in a specified time period. The general value of α in engineering application is 0.05, which means that the correct probability is 95% when the determination of the original assumption is made, and α can be 0.01 if the accuracy requirement on the detection result is high. In the embodiment, when alpha is 0.05, and the test statistic meets |Z| not less than Z 1-α/2, namely |Z| not less than Z 0.975 =1.96, the original assumption is accepted, and the current parameter is considered to have abnormal running trend in the parameter data set X to be tested, which is acquired in a specified time period; and if the test statistic meets |Z| < Z 1-α/2, namely |Z| < Z 0.975 =1.96, rejecting the original assumption, and considering that the running trend abnormality does not exist in the parameter data set X to be tested, which is acquired by the current parameter in a specified time period.
In the steps 4.1 to 4.5, the variance of the operation parameter to be detected is calculated to determine whether an abnormality occurs, and of course, other conventional detection methods may be adopted to determine whether the operation parameter to be detected has an abnormality.
And 5, taking the normal operation parameters subjected to screening and pretreatment in the step 2 as samples, and establishing a mapping connection relation network of the SBW integrated wireless temperature sensor by using a Granger causality examination method, wherein the mapping connection relation network is shown in fig. 3. The mapping connection relation network is used for determining the coupling relation between each parameter of the target temperature sensor and each component of each module, and when the module or the component has abnormal operation trend, the parameter data with the coupling connection relation has abnormal conditions with different degrees.
And 5.1, dividing the normal operation parameter sample subjected to screening and pretreatment in the step 2 into a strong state characterization parameter set and a weak state characterization parameter set, which are respectively marked as P S＝(ps1,ps2,…,psm) and P W＝(pw1,pw2,…,pwn). The classification standard of the strong state characterization parameter set is as follows: the parameters and the modules have direct and unique physical connection relation, and the state information of the modules can be directly represented. If the ratio of the resistance value of the thermal resistor to the actual temperature value of the measured environment is the same, the thermal resistor component used for measuring the temperature of the SBW integrated wireless temperature sensor and the temperature sensing module have direct and unique physical connection relation, namely the parameter only belongs to the current module, and can directly represent all state information of the thermal resistor and part of state information of the temperature sensing module; the classification criteria of the weak state characterization parameter set are: the parameters and the modules have indirect or non-unique physical connection relation, and the data value can represent part of state information of the whole SBW integrated wireless temperature sensor, but the attribution of the specific module cannot be clarified. For example, the loop current of the circuit part of the SBW integrated wireless temperature sensor has a physical connection relation with each module of the SBW integrated wireless temperature sensor, but the specific membership relation is not clear, and the abnormality of the module or the component cannot be determined to be caused by the abnormality of the module or the component, or the abnormality of the module or the component can cause the abnormality of the module or the component.
And 5.2, establishing a mapping connection relation network of the SBW integrated wireless temperature sensor.
Setting each parameter P sm in the strong state characterization parameter set P S as a target parameter, and checking each parameter P wn in the weak state characterization parameter set P W by using a Granger causal relationship checking method, wherein each parameter P sm in the strong state characterization parameter set P S and each module of the SBW integrated wireless temperature sensor have a direct and unique physical connection relationship, and if a certain parameter P wk in the weak state characterization parameter set P W and a certain parameter P sj in the strong state characterization parameter set P S have a causal connection relationship, the parameters are correspondingly subordinate to the module directly corresponding to P sj in the constructed mapping connection relationship network. According to the step of Granger causal relation test method, the following regression model is established for all the strong state characterization parameters p sj in the normal operation parameter data set and the weak state characterization parameters p wk in the normal operation parameter data set:
Wherein, (1) is a constrained regression model, (2) is an unconstrained regression model, u and v are the hysteresis order of p sj and p wk, respectively, and the test zero hypothesis is: p wk is not the Granger cause of the change in p sj, i.e. H 0:β1,β2,…,βv =0. Constrained residual square sum RSS s is obtained through (1) formula calculation, unconstrained residual square sum RSS s|w is obtained through (2) formula calculation, and test statistics are constructed It obeys the F distribution with the molecular degree of freedom of v denominator degree of freedom of n-u-v-1. Taking the significance level alpha=0.05, if F > F 0.05, rejecting the zero hypothesis, and considering p wk as a Granger cause for p sj change, namely, p wk belongs to a module or a component directly corresponding to p sj in the constructed mapping connection relation network, wherein the two modules or components have a connection relation.
And 5.3, performing reverse Granger causal relationship test on all the strong state characterization parameters p sj and the weak state characterization parameters p wk in a manner of following the step 5.2, namely checking whether p sj is a Granger cause causing the change of p wk, and constructing a complete SBW integrated wireless temperature sensor parameter-component-module-system mapping connection relation network.
And 5.4, determining the importance degree of each parameter in the connection module, namely, mapping the weight coefficient of each connection relation in the connection relation network. If the ratio of the resistance value of the temperature sensing module to the actual temperature value of the measured environment is the thermal resistance used for measuring the temperature, the indirect correlation module is the temperature sensing module, and the parameters corresponding to the thermal resistance part of the SBW integrated wireless temperature sensor are only the ratio, so that the importance degree of the ratio in the thermal resistance part is 100%, the weight coefficient is 1, and when the abnormal operation trend of the ratio of the resistance value to the actual temperature value of the measured environment is detected, the abnormal operation condition of the thermal resistance part of the SBW integrated wireless temperature sensor is considered; the above ratio is one of all parameters corresponding to the temperature sensing module of the SBW integrated wireless temperature sensor, when the above ratio is abnormal in operation trend, the temperature sensing module is not necessarily abnormal in operation trend, and the importance degree of the temperature sensing module is determined after a statistical test, for example, 100 times of abnormal injection tests are performed, wherein 63 times of abnormal operation trend is detected in the temperature sensing module, and the importance degree of the above ratio in the temperature sensing module is approximately 63% and the importance degree coefficient is 0.63.
And step 6, sequentially executing the step 4 on the operation parameters to be detected to obtain the abnormal occurrence condition of the operation trend of each parameter. And then according to the evaluation system established in the step 5, the operation trend abnormality judgment is carried out on the SBW integrated wireless temperature sensor, and the specific method comprises the following steps:
Step 6.1, when the industrial temperature sensor is in any single temperature state, the operation parameters to be detected are connected with the relation network through mapping, and if the probability of failure of a specific module part is greater than a first threshold value, the module part is directly judged to be failed;
Step 6.2, when the industrial temperature sensor is in any single temperature state, the operation parameters to be detected are mapped to a connection relation network through a mapping, the probability of failure of a specific module part is smaller than a first threshold value and larger than a second threshold value, and when the probability of failure is larger than the second threshold value, comprehensive judgment is carried out by combining the failure probabilities in the other two temperature states;
And 6.3, when the industrial temperature sensor is in any single temperature state, judging that the operation parameters to be detected are normal through the mapping connection relation network and when the probability of failure of the specific module component is smaller than a second threshold value.
The specific method for comprehensively judging by combining the fault probabilities in the other two temperature states comprises the following steps:
And calculating the average probability of failure of the specific module component under three temperature states, and judging that the module component fails if the average probability is larger than a preset value.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations 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.
Finally, it should be noted that: the above embodiments are only for illustrating the technical aspects of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the above embodiments, it should be understood by those of ordinary skill in the art that: modifications and equivalents may be made to the specific embodiments of the invention without departing from the spirit and scope of the invention, which is intended to be covered by the claims.
Claims (5)
1. The abnormality detection method of the industrial temperature sensor in the large temperature difference environment is characterized by comprising the following steps of:
collecting operation parameters to be detected of an industrial temperature sensor to be detected in a high-temperature and low-temperature variable-temperature state;
If the operation parameters to be detected are abnormal, comprehensively deciding the operation parameters to be detected in the high-temperature, low-temperature and variable-temperature states through a mapping connection relation network, and judging the abnormal probability of each module component of the industrial temperature sensor;
the mapping connection relation network is used for mapping the fault probability relation between each operation parameter of the industrial temperature sensor in the high-temperature, low-temperature and variable-temperature states and the corresponding module component;
the method for judging the operation abnormality of the operation parameters to be detected comprises the following steps:
recording normal operation parameters of the industrial temperature sensor in a normal working state under high temperature and low temperature and variable temperature states, and calculating normal operation mathematical expectations of the parameters;
carrying out hypothesis testing analysis on the operation parameters to be detected and normal operation mathematical expectations in sequence, and judging whether the industrial temperature sensor is abnormal or not;
The method for establishing the mapping connection relation network comprises the following steps:
The repeated abnormal injection test is carried out on each module component of the industrial temperature sensor under the conditions of high temperature, low temperature and variable temperature respectively;
under the conditions of high temperature and low temperature and variable temperature, counting the abnormal occurrence times and degree of each abnormal operation parameter with a mapping relation, establishing a mapping relation between each operation parameter and each module component of the industrial temperature sensor by using a Granger causal relation test method according to the ratio of the statistical data, and determining the abnormal weight coefficient of each abnormal operation parameter and the corresponding module component under different temperature conditions;
The method for judging the abnormal probability of each module component of the industrial temperature sensor comprises the following specific steps:
when the industrial temperature sensor is in any single temperature state, the operation parameter to be detected passes through a mapping connection relation network, and the probability of failure of a specific module part is greater than a first threshold value, and the module part is directly judged to be failed;
When the industrial temperature sensor is in any single temperature state, the operation parameters to be detected are connected with the relation network through mapping, the probability of failure of the specific module part is smaller than a first threshold value, and when the probability of failure of the specific module part is larger than a second threshold value, the comprehensive judgment is carried out by combining the failure probabilities in the other two temperature states;
when the industrial temperature sensor is in any single temperature state, the operation parameter to be detected passes through the mapping connection relation network, and when the probability of failure of the specific module component is smaller than a second threshold value, the operation parameter is judged to be normal;
The specific method for comprehensively judging by combining the fault probabilities in the other two temperature states comprises the following steps:
Calculating the average probability of failure of a specific module component in three temperature states, and judging that the module component fails if the average probability is larger than a preset value;
The specific method for judging whether the industrial temperature sensor is abnormal or not by carrying out hypothesis test analysis on the operation parameters to be detected and the normal operation mathematical expectation in sequence comprises the following steps:
the parameters are subjected to time sequence difference operation, difference normalization summation, difference sum variance operation, and test statistics and double-tail test.
2. The method for detecting the abnormality of the industrial temperature sensor in the environment with the large temperature difference according to claim 1, wherein the industrial temperature sensor comprises a temperature sensing module, a temperature transmitting module, a digital display module, a wireless transmitting module and a main control module.
3. The method for detecting an abnormality of an industrial temperature sensor in a large temperature difference environment according to claim 2, wherein the operation parameters include:
resistance value, current value and voltage value when the temperature sensing module operates;
The temperature transmitting module is used for transmitting an input signal value and an output signal value when in operation;
the working voltage and the battery power of the digital display module during operation;
signal intensity, capacitance value and inductance value when the wireless transmitting module operates;
motherboard temperature and working voltage when the main control module operates.
4. An abnormality detection system for an industrial temperature sensor in a large temperature difference environment using the detection method according to claim 1, characterized in that the system comprises:
the acquisition unit is used for acquiring operation parameters to be detected of the industrial temperature sensor to be detected in a high-temperature, low-temperature and variable-temperature state;
the abnormality judging unit is used for judging whether the operation parameters to be detected are abnormal;
And the mapping connection relation network unit is used for comprehensively deciding the operation parameters to be detected in the high-temperature and low-temperature variable-temperature states if the operation parameters to be detected are abnormal, and judging the abnormal probability of each module component of the industrial temperature sensor.
5. A storage medium storing instructions adapted to be loaded by a processor to perform the method of any one of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310261906.8A CN116295948B (en) | 2023-03-17 | 2023-03-17 | Abnormality detection method, system and storage medium of industrial temperature sensor in large temperature difference environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310261906.8A CN116295948B (en) | 2023-03-17 | 2023-03-17 | Abnormality detection method, system and storage medium of industrial temperature sensor in large temperature difference environment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116295948A CN116295948A (en) | 2023-06-23 |
| CN116295948B true CN116295948B (en) | 2024-07-19 |
Family
ID=86800984
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310261906.8A Active CN116295948B (en) | 2023-03-17 | 2023-03-17 | Abnormality detection method, system and storage medium of industrial temperature sensor in large temperature difference environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116295948B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117249921B (en) * | 2023-11-15 | 2024-04-05 | 宁德时代新能源科技股份有限公司 | Abnormality identification method for temperature sampling, related device, vehicle and storage medium |
| CN117308275B (en) * | 2023-11-28 | 2024-02-06 | 江苏嘉尚环保科技有限公司 | Temperature difference-based pipeline connection abnormality detection method and system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109990923A (en) * | 2016-06-23 | 2019-07-09 | 福州丹诺西诚电子科技有限公司 | Improve the fault diagnosis method and system of the temperature sensor of utilization of resources rate |
| CN112834079A (en) * | 2020-12-25 | 2021-05-25 | 山东朗进科技股份有限公司 | Method for judging parameter drift of temperature sensor of air conditioning unit of railway vehicle |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7991524B2 (en) * | 2008-01-02 | 2011-08-02 | GM Global Technology Operations LLC | Temperature sensor diagnostics |
| JP5411998B1 (en) * | 2012-12-28 | 2014-02-12 | 富士重工業株式会社 | Temperature sensor diagnostic device |
| CN112229543A (en) * | 2020-10-23 | 2021-01-15 | 蔚来汽车科技(安徽)有限公司 | Method and apparatus for determining state of battery temperature sensor, medium, vehicle, and server |
-
2023
- 2023-03-17 CN CN202310261906.8A patent/CN116295948B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109990923A (en) * | 2016-06-23 | 2019-07-09 | 福州丹诺西诚电子科技有限公司 | Improve the fault diagnosis method and system of the temperature sensor of utilization of resources rate |
| CN112834079A (en) * | 2020-12-25 | 2021-05-25 | 山东朗进科技股份有限公司 | Method for judging parameter drift of temperature sensor of air conditioning unit of railway vehicle |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116295948A (en) | 2023-06-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116295948B (en) | Abnormality detection method, system and storage medium of industrial temperature sensor in large temperature difference environment | |
| CN111459700B (en) | Equipment fault diagnosis method, diagnosis device, diagnosis equipment and storage medium | |
| CN112284440B (en) | Sensor data deviation self-adaptive correction method | |
| CN106557546B (en) | Method and system for mining and evaluating extra-high voltage online monitoring data | |
| CN113642754A (en) | Complex industrial process fault prediction method based on RF noise reduction self-coding information reconstruction and time convolution network | |
| CN109492790A (en) | Wind turbines health control method based on neural network and data mining | |
| CN107942994A (en) | A kind of satellite temperature control system method for diagnosing faults based on temperature curve feature | |
| CN115455746B (en) | Nuclear power device operation monitoring data anomaly detection and correction integrated method | |
| CN115392812B (en) | Abnormal root cause positioning method, device, equipment and medium | |
| CN111931334A (en) | Method and system for evaluating operation reliability of cable equipment | |
| KR20220068799A (en) | System for detecting error of automation equipment and method thereof | |
| CN112580858A (en) | Equipment parameter prediction analysis method and system | |
| CN107103425B (en) | Intelligent energy evaluation system for power generation equipment running state computer | |
| CN114597886A (en) | Power distribution network operation state evaluation method based on interval type two fuzzy clustering analysis | |
| CN117388703A (en) | Capacitor aging state evaluation method based on improved depth residual error | |
| CN112326246A (en) | Bearing safety state online monitoring method based on periodic data and nuclear density estimation | |
| CN116910677A (en) | Industrial instrument fault diagnosis method and system | |
| CN117077051A (en) | Self-adaptive identification method for dam monitoring abnormal data | |
| CN114167837B (en) | Intelligent fault diagnosis method and system for railway signal system | |
| CN111614504A (en) | Power grid regulation and control data center service characteristic fault positioning method and system based on time sequence and fault tree analysis | |
| CN110889646A (en) | Method for checking correctness of operating parameters of generator set adapting to working condition change | |
| CN116502526A (en) | Improved PSO-GRNN neural network-based weighing sensor fault diagnosis method | |
| CN116451121A (en) | Switch fault diagnosis method, system, equipment and medium | |
| CN114400776A (en) | Substation automation equipment state diagnosis method and system based on digital mirror image | |
| CN114580101B (en) | Method and system for predicting residual service life of rotary machine |
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 |