CN113588123A - Stator winding temperature early warning method - Google Patents
Stator winding temperature early warning method Download PDFInfo
- Publication number
- CN113588123A CN113588123A CN202110865263.9A CN202110865263A CN113588123A CN 113588123 A CN113588123 A CN 113588123A CN 202110865263 A CN202110865263 A CN 202110865263A CN 113588123 A CN113588123 A CN 113588123A
- Authority
- CN
- China
- Prior art keywords
- stator winding
- winding temperature
- value
- measuring point
- temperature measuring
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
- G01K13/00—Thermometers specially adapted for specific purposes
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
- G01K1/00—Details of thermometers not specially adapted for particular types of thermometer
- G01K1/02—Means for indicating or recording specially adapted for thermometers
- G01K1/024—Means for indicating or recording specially adapted for thermometers for remote indication
Abstract
The invention discloses a stator winding temperature early warning method, which belongs to the field of generators and is characterized by comprising the following steps of: s1, setting a threshold value K, a total judgment interval N and a total early warning N0(ii) a S2, calculating theta't(ii) a S3, if theta'tAnd thetatWithin K, determining thetatIf the counter is normal, the counter is increased by 1, and the abnormal counter is unchanged; s4, if theta'tAnd thetatExceeds K, and theta is determinedtIf the counter is abnormal, the counter is increased by 1, and the abnormal counter is increased by 1; s5, if the value of the abnormal counter is less than N0And judging that the value of the counter is less than N, and jumping to step S2; s6, setting the abnormal calculator and the judgment counter to zero; s7, if the value of the abnormal counter is larger than or equal to N0And then sending out an early warning signal. The invention can identify abnormal temperature data, find early thermal fault, combine with counter to count, when the number of overrun accumulates a certain value, send out the abnormityAnd the alarm is given, and the temperature early warning is more accurate.
Description
Technical Field
The invention relates to the technical field of generators, in particular to a stator winding temperature early warning method.
Background
The stator winding of the large-scale generator is placed in a slot of a stator core, and the straight line part is positioned in a rotating main magnetic field to induce high voltage and large current to be transmitted to a power grid. The stator winding is used as a key component for energy conversion and electric energy output of the generator, and the quality of the running state of the stator winding directly influences whether the whole unit can run safely and stably. Because the stator winding of a large generator has large current, the stator current of the steam turbine generator with the power of 300MW and 600MW respectively exceeds 10000A and 20000A, and therefore the stator winding is one of the parts with the largest loss and heat generation of the generator. Statistical data show that the stator thermal fault is a common fault of the generator, and because the temperature of the stator winding is a key sign of the fault, each power plant pays extra attention to the temperature of the stator winding.
At present, a fixed limit value alarm mechanism is generally adopted by a power plant, namely, a temperature limit value is set, once data of temperature measuring points arranged on a stator winding exceed the value, an alarm signal is sent out to remind monitoring personnel of the power plant to confirm and process, the temperature limit alarm value is generally set according to design or related standards, and for a large-sized internal water cooling steam turbine generator, the temperature alarm value of a water outlet end of the stator winding is 85 ℃, and the temperature alarm value in a tank is 90 ℃. The alarm mechanism alarms when the stator winding of the generator has obvious faults and reaches the limit, however, in order to meet the requirements of a power grid, the operation mode of a large-scale generator set is more flexible than the prior art, the peak regulation is frequently and deeply performed, the stator current is far lower than the rated current after long-time low-load operation, correspondingly, the temperature of the water outlet end of the stator winding and the temperature in the groove are much lower than the normal rated working condition, under the condition, when the early thermal fault occurs but does not exceed the limit, the monitoring function of fixed limit alarm is greatly weakened, and the early symptoms of the thermal fault of the stator cannot be effectively and timely found.
Chinese patent literature with publication number CN 108847799A and publication date of 2018, 11 and 20 discloses a PMSM stator winding temperature online detection method based on signal injection, which is characterized by comprising the following steps: establishing a real-time temperature observation method for a stator winding of a permanent magnet synchronous motor; and step two, adding an optimal injection signal strategy into the temperature observation method in the step one.
The PMSM stator winding temperature online detection method based on signal injection disclosed by the patent document can monitor the health condition of a motor and prevent over-temperature by estimating the stator winding temperature of a permanent magnet synchronous motor online, can also be used in the optimization control of an active heat management motor, and is beneficial to improving the performance of an electric drive system. However, abnormal temperature data cannot be identified in time, and the early warning accuracy is low.
Disclosure of Invention
The invention provides a stator winding temperature early warning method for overcoming the defects of the prior art, and the method is based on a self-evaluation alarm mechanism of temperature online calculation, takes a theoretical normal value of dynamically changing temperature as a judgment standard, the judgment standard is closely related to the running condition of a generator, abnormal temperature data can be timely identified, early thermal faults are found, the counting of a counter is combined, when a certain value is accumulated by the number of times of exceeding limit, an abnormal alarm is sent, the temperature data drifts under the non-abnormal condition, a certain fault tolerance capability is realized, and the temperature early warning is more accurate.
The invention is realized by the following technical scheme:
a stator winding temperature early warning method is characterized by comprising the following steps:
s1, setting an identification convergence error value and a ratio value, identifying a coefficient vector required by the on-line calculation of the stator winding temperature, and setting a threshold value K, a judgment interval total number N and an early warning total number N0;
S2, extracting n determinantsAnd (3) calculating a continuous operation value of the sub-winding temperature, and calculating an operation predicted value theta 'of a stator winding temperature measuring point at the time t according to the coefficient vector by using formula 1't;
Wherein, theta'tFor the stator winding temperature measuring point operation predicted value at the time t, thetat-1,θt-2…θt-nFor measuring actual running value, alpha, of stator winding temperature before time t1,α2…αnIn order to weight the vector of coefficients,as disturbance quantity, αiAs a vector of weighting coefficients, thetat-iThe actual operation value of the stator winding temperature measuring point at the time t-i is obtained;
s3, if the operation predicted value theta of the stator winding temperature measuring point at the moment t'tActual operation value theta of stator winding temperature measuring point at t momenttWhen the deviation is within the threshold value K, the actual operation value theta of the stator winding temperature measuring point at the time t is judgedtAnd (4) if the counter is judged to be added by 1 and the abnormal counter is not changed, the operation predicted value theta 'of the stator winding temperature measuring point at the moment t + 1't+1The calculation is determined by formula 2;
wherein, theta't+1The predicted value alpha of the operation of the stator winding temperature measuring point at the moment t +1iAs a vector of weighting coefficients, thetat+1-iIs the actual operation value of the stator winding temperature measuring point at the time t +1-i,is the disturbance quantity;
s4, if the operation predicted value theta of the stator winding temperature measuring point at the moment t'tActual operation value theta of stator winding temperature measuring point at t momenttWhen the deviation exceeds a threshold value K, the actual operation value theta of the stator winding temperature measuring point at the moment t is judgedtIf the abnormal data does not meet the normal characteristic, the abnormal data is judged, the counter is increased by 1, the abnormal counter is increased by 1, and the actual operation value theta of the stator winding temperature measuring point at the time t is eliminatedtAnd if the stator winding temperature measuring point does not participate in subsequent calculation, the operation predicted value theta 'of the stator winding temperature measuring point at the moment t + 1't+1The calculation is determined by formula 3;
wherein, theta't+1The predicted value alpha of the operation of the stator winding temperature measuring point at the moment t +1iAs a vector of weighting coefficients, thetat-iIs the actual operation value of the stator winding temperature measuring point at the time t-i,is the disturbance quantity;
s5, if the value of the abnormal counter is less than the total number N of the early warnings0And the value of the judgment counter is smaller than the total number N of the judgment intervals, and the step is shifted to S2 after moving a moment;
s6, if the value of the abnormal counter is less than the total number N of the early warnings0If the value of the judgment counter is larger than or equal to the total number N of the judgment intervals, the actual operation values of the stator winding temperature measuring points of the total number N of the judgment intervals are considered to be normal, the abnormality calculator and the judgment counter are set to zero, the operation is moved backwards by one moment, the step S2 is skipped, and the next round of judgment is started;
s7, if the value of the abnormal counter is larger than or equal to the total number N of the early warnings0And if the actual operation value of the stator winding temperature measuring point in the interval contained by the judgment counter is in a degradation trend, sending out an early warning signal, setting the abnormality calculator and the judgment counter to be zero, moving backwards for one moment, skipping to the step S2, and starting the next round of judgment.
In step S1, identifying the coefficient vector required for the on-line calculation of the stator winding temperature specifically refers to selecting a total n value of the coefficient vector, selecting continuous operation data of the stator winding temperature measurement points for a period of time, setting an iteration initial value of the coefficient vector, identifying the coefficient vector by a least square method or a neural network, judging whether a ratio of the number of errors smaller than a convergence error value to the total number reaches the standard after the calculation error is stable in the statistical identification process, if not, reselecting the total n value of the coefficient vector, and adjusting the parameter to continue the identification; if yes, the coefficient vector is output.
Running predicted value theta 'of stator winding temperature measuring point at moment t'tActual operation value theta of stator winding temperature measuring point at t momenttThe deviation of (2) is absolute deviation sigma, and the absolute deviation sigma is calculated by formula 4;
σ=|θt-θ'tequation 4
Where σ is the absolute deviation, θtIs an actual operation value theta of a stator winding temperature measuring point at the time t'tAnd (5) running a predicted value for the temperature measuring point of the stator winding at the time t.
Running predicted value theta 'of stator winding temperature measuring point at moment t'tActual operation value theta of stator winding temperature measuring point at t momenttThe deviation of (a) is calculated by using a relative deviation lambda which is calculated by formula 5;
λ=|θt-θ't|/θtformula 5
Where λ is the relative deviation, θtIs an actual operation value theta of a stator winding temperature measuring point at the time t'tAnd (5) running a predicted value for the temperature measuring point of the stator winding at the time t.
The actual operation value theta of the stator winding temperature measuring point at the time of the threshold value K and the time ttThe actual operation value theta of the stator winding temperature measuring point at the time t in inverse proportiontThe larger the threshold K, the smaller the threshold K.
The total number of early warnings N0Less than or equal to the total number of judgment intervals N.
The basic principle of the invention is as follows:
the theoretical normal temperature value of each measuring point of the stator winding under any load is obtained through research, the theoretical normal temperature value is obtained through deep research and repeated verification on the mechanism and the structure of the generator, then the running condition of the generator is evaluated in real time through comparison of the actually measured running value of the generator and the theoretical normal temperature value, counting is carried out when the running data of the measuring points of the unit deviates from a certain threshold value compared with the theoretical normal temperature value, and an alarm is sent out when the counting reaches a certain value within a certain time.
The beneficial effects of the invention are mainly shown in the following aspects:
compared with the fixed limit value alarm, the invention has the advantages that the self-evaluation alarm mechanism based on the temperature on-line calculation takes the theoretical normal value of the dynamically changing temperature as the judgment standard, the judgment standard is closely related to the running condition of the generator, the abnormal temperature data can be identified in time, the early thermal fault can be found, the abnormal alarm can be sent out by combining the counting of the counter when the number of times of exceeding limit is accumulated to a certain value, the temperature data drifts under the non-abnormal condition, the self-evaluation alarm mechanism has certain fault tolerance capability, and the temperature early warning is more accurate.
According to the invention, according to the running temperature characteristics of the stator winding, the local normal running characteristics of the stator winding temperature measuring point can be described only by introducing the recent temperature running value and a group of coefficient vectors, and the local temperature of the stator winding is obtained by real-time online calculation instead of the average temperature of the stator winding, so that the validity of the health evaluation of the generator is better.
The method is suitable for different load working conditions of flexible operation of the generator, and can be used for respectively modeling each temperature measuring point of the stator winding and extracting coefficient vectors which accord with respective operation characteristics, so that a high-precision calculation model is constructed, and the calculation precision is ensured.
According to the invention, by introducing the normal operation value of the temperature at the previous moment, a high-frequency signal is not required to be injected, the safe operation risk of the generator is avoided, the method is suitable for different load working conditions, the adaptability to each temperature measuring point of the stator winding is high, the individualized operation characteristics of each measuring point can be met, the local temperature of the stator winding can be calculated in real time on line, the calculation precision is high, and the effect of evaluating the health of the generator is greatly improved.
Drawings
The invention will be further described in detail with reference to the drawings and the detailed description, wherein:
FIG. 1 is a block flow diagram of the present invention.
Detailed Description
Example 1
Referring to fig. 1, a stator winding temperature early warning method includes the following steps:
s1, setting an identification convergence error value and a ratio value, identifying a coefficient vector required by the on-line calculation of the stator winding temperature, and setting a threshold value K, a judgment interval total number N and an early warning total number N0;
S2, extracting n stator winding temperature continuous operation values, and calculating a stator winding temperature measuring point operation predicted value theta 'at t moment according to the coefficient vector through formula 1't;
Wherein, theta'tFor the stator winding temperature measuring point operation predicted value at the time t, thetat-1,θt-2…θt-nFor measuring actual running value, alpha, of stator winding temperature before time t1,α2…αnIn order to weight the vector of coefficients,as disturbance quantity, αiAs a vector of weighting coefficients, thetat-iThe actual operation value of the stator winding temperature measuring point at the time t-i is obtained;
s3, if the operation predicted value theta of the stator winding temperature measuring point at the moment t'tActual operation value theta of stator winding temperature measuring point at t momenttWhen the deviation is within the threshold value K, the actual operation value theta of the stator winding temperature measuring point at the time t is judgedtAnd (4) if the counter is judged to be added by 1 and the abnormal counter is not changed, the operation predicted value theta 'of the stator winding temperature measuring point at the moment t + 1't+1The calculation is determined by formula 2;
wherein, theta't+1The predicted value alpha of the operation of the stator winding temperature measuring point at the moment t +1iAs a vector of weighting coefficients, thetat+1-iIs the actual operation value of the stator winding temperature measuring point at the time t +1-i,is the disturbance quantity;
s4, if the operation predicted value theta of the stator winding temperature measuring point at the moment t'tActual operation value theta of stator winding temperature measuring point at t momenttWhen the deviation exceeds a threshold value K, the actual operation value theta of the stator winding temperature measuring point at the moment t is judgedtIf the abnormal data does not meet the normal characteristic, the abnormal data is judged, the counter is increased by 1, the abnormal counter is increased by 1, and the actual operation value theta of the stator winding temperature measuring point at the time t is eliminatedtAnd if the stator winding temperature measuring point does not participate in subsequent calculation, the operation predicted value theta 'of the stator winding temperature measuring point at the moment t + 1't+1The calculation is determined by formula 3;
wherein, theta't+1The predicted value alpha of the operation of the stator winding temperature measuring point at the moment t +1iAs a vector of weighting coefficients, thetat-iIs the actual operation value of the stator winding temperature measuring point at the time t-i,is the disturbance quantity;
s5, if the value of the abnormal counter is less than the total number N of the early warnings0And the value of the judgment counter is smaller than the total number N of the judgment intervals, and the step is shifted to S2 after moving a moment;
s6, if the value of the abnormal counter is less than the total number N of the early warnings0And if the value of the judgment counter is larger than or equal to the total number N of the judgment intervals, the actual operation values of the stator winding temperature measuring points of the total number N of the judgment intervals are considered to be normal, the abnormality calculator and the judgment counter are all set to zero, and the abnormality calculator and the judgment counter move backwards for one momentSkipping to step S2, and starting the next round of judgment;
s7, if the value of the abnormal counter is larger than or equal to the total number N of the early warnings0And if the actual operation value of the stator winding temperature measuring point in the interval contained by the judgment counter is in a degradation trend, sending out an early warning signal, setting the abnormality calculator and the judgment counter to be zero, moving backwards for one moment, skipping to the step S2, and starting the next round of judgment.
The embodiment is the most basic implementation mode, compared with the alarm with a fixed limit value, the self-evaluation alarm mechanism based on the temperature on-line calculation takes a theoretical normal value of the dynamically changing temperature as a judgment standard, the judgment standard is closely related to the operation condition of the generator, abnormal temperature data can be identified in time, early thermal faults can be found, the counter is combined for counting, when a certain value is accumulated by the number of times of overrun, the abnormal alarm is sent out, the temperature data drifts under the non-abnormal condition, certain fault tolerance is achieved, and the temperature early warning is more accurate.
Example 2
Referring to fig. 1, a stator winding temperature early warning method includes the following steps:
s1, setting an identification convergence error value and a ratio value, identifying a coefficient vector required by the on-line calculation of the stator winding temperature, and setting a threshold value K, a judgment interval total number N and an early warning total number N0;
S2, extracting n stator winding temperature continuous operation values, and calculating a stator winding temperature measuring point operation predicted value theta 'at t moment according to the coefficient vector through formula 1't;
Wherein, theta'tFor the stator winding temperature measuring point operation predicted value at the time t, thetat-1,θt-2…θt-nFor measuring actual running value, alpha, of stator winding temperature before time t1,α2…αnIn order to weight the vector of coefficients,as disturbance quantity, αiAs a vector of weighting coefficients, thetat-iThe actual operation value of the stator winding temperature measuring point at the time t-i is obtained;
s3, if the operation predicted value theta of the stator winding temperature measuring point at the moment t'tActual operation value theta of stator winding temperature measuring point at t momenttWhen the deviation is within the threshold value K, the actual operation value theta of the stator winding temperature measuring point at the time t is judgedtAnd (4) if the counter is judged to be added by 1 and the abnormal counter is not changed, the operation predicted value theta 'of the stator winding temperature measuring point at the moment t + 1't+1The calculation is determined by formula 2;
wherein, theta't+1The predicted value alpha of the operation of the stator winding temperature measuring point at the moment t +1iAs a vector of weighting coefficients, thetat+1-iIs the actual operation value of the stator winding temperature measuring point at the time t +1-i,is the disturbance quantity;
s4, if the operation predicted value theta of the stator winding temperature measuring point at the moment t'tActual operation value theta of stator winding temperature measuring point at t momenttWhen the deviation exceeds a threshold value K, the actual operation value theta of the stator winding temperature measuring point at the moment t is judgedtIf the abnormal data does not meet the normal characteristic, the abnormal data is judged, the counter is increased by 1, the abnormal counter is increased by 1, and the actual operation value theta of the stator winding temperature measuring point at the time t is eliminatedtAnd if the stator winding temperature measuring point does not participate in subsequent calculation, the operation predicted value theta 'of the stator winding temperature measuring point at the moment t + 1't+1The calculation is determined by formula 3;
wherein, theta't+1For the stator winding temperature measuring point operation prediction at the moment of t +1Measured value, αiAs a vector of weighting coefficients, thetat-iIs the actual operation value of the stator winding temperature measuring point at the time t-i,is the disturbance quantity;
s5, if the value of the abnormal counter is less than the total number N of the early warnings0And the value of the judgment counter is smaller than the total number N of the judgment intervals, and the step is shifted to S2 after moving a moment;
s6, if the value of the abnormal counter is less than the total number N of the early warnings0If the value of the judgment counter is larger than or equal to the total number N of the judgment intervals, the actual operation values of the stator winding temperature measuring points of the total number N of the judgment intervals are considered to be normal, the abnormality calculator and the judgment counter are set to zero, the operation is moved backwards by one moment, the step S2 is skipped, and the next round of judgment is started;
s7, if the value of the abnormal counter is larger than or equal to the total number N of the early warnings0And if the actual operation value of the stator winding temperature measuring point in the interval contained by the judgment counter is in a degradation trend, sending out an early warning signal, setting the abnormality calculator and the judgment counter to be zero, moving backwards for one moment, skipping to the step S2, and starting the next round of judgment.
In step S1, identifying the coefficient vector required for the on-line calculation of the stator winding temperature specifically refers to selecting a total n value of the coefficient vector, selecting continuous operation data of the stator winding temperature measurement points for a period of time, setting an iteration initial value of the coefficient vector, identifying the coefficient vector by a least square method or a neural network, judging whether a ratio of the number of errors smaller than a convergence error value to the total number reaches the standard after the calculation error is stable in the statistical identification process, if not, reselecting the total n value of the coefficient vector, and adjusting the parameter to continue the identification; if yes, the coefficient vector is output.
According to the operating temperature characteristics of the stator winding, the local normal operating characteristics of the stator winding temperature measuring points can be described only by introducing recent temperature operating values and a group of coefficient vectors, and the local temperature of the stator winding is obtained through real-time online calculation instead of the average temperature of the stator winding, so that the effectiveness of the health evaluation of the generator is better.
Example 3
Referring to fig. 1, a stator winding temperature early warning method includes the following steps:
s1, setting an identification convergence error value and a ratio value, identifying a coefficient vector required by the on-line calculation of the stator winding temperature, and setting a threshold value K, a judgment interval total number N and an early warning total number N0;
S2, extracting n stator winding temperature continuous operation values, and calculating a stator winding temperature measuring point operation predicted value theta 'at t moment according to the coefficient vector through formula 1't;
Wherein, theta'tFor the stator winding temperature measuring point operation predicted value at the time t, thetat-1,θt-2…θt-nFor measuring actual running value, alpha, of stator winding temperature before time t1,α2…αnIn order to weight the vector of coefficients,as disturbance quantity, αiAs a vector of weighting coefficients, thetat-iThe actual operation value of the stator winding temperature measuring point at the time t-i is obtained;
s3, if the operation predicted value theta of the stator winding temperature measuring point at the moment t'tActual operation value theta of stator winding temperature measuring point at t momenttWhen the deviation is within the threshold value K, the actual operation value theta of the stator winding temperature measuring point at the time t is judgedtAnd (4) if the counter is judged to be added by 1 and the abnormal counter is not changed, the operation predicted value theta 'of the stator winding temperature measuring point at the moment t + 1't+1The calculation is determined by formula 2;
wherein, theta't+1At time t +1Stator winding temperature measurement point operation prediction value, alphaiAs a vector of weighting coefficients, thetat+1-iIs the actual operation value of the stator winding temperature measuring point at the time t +1-i,is the disturbance quantity;
s4, if the operation predicted value theta of the stator winding temperature measuring point at the moment t'tActual operation value theta of stator winding temperature measuring point at t momenttWhen the deviation exceeds a threshold value K, the actual operation value theta of the stator winding temperature measuring point at the moment t is judgedtIf the abnormal data does not meet the normal characteristic, the abnormal data is judged, the counter is increased by 1, the abnormal counter is increased by 1, and the actual operation value theta of the stator winding temperature measuring point at the time t is eliminatedtAnd if the stator winding temperature measuring point does not participate in subsequent calculation, the operation predicted value theta 'of the stator winding temperature measuring point at the moment t + 1't+1The calculation is determined by formula 3;
wherein, theta't+1The predicted value alpha of the operation of the stator winding temperature measuring point at the moment t +1iAs a vector of weighting coefficients, thetat-iIs the actual operation value of the stator winding temperature measuring point at the time t-i,is the disturbance quantity;
s5, if the value of the abnormal counter is less than the total number N of the early warnings0And the value of the judgment counter is smaller than the total number N of the judgment intervals, and the step is shifted to S2 after moving a moment;
s6, if the value of the abnormal counter is less than the total number N of the early warnings0If the value of the judgment counter is larger than or equal to the total number N of the judgment intervals, the actual operation values of the stator winding temperature measuring points of the total number N of the judgment intervals are considered to be normal, the abnormality calculator and the judgment counter are set to zero, the operation is moved backwards by one moment, the step S2 is skipped, and the next round of judgment is started;
s7, if the value of the abnormal counter is largeIs equal to or more than the total number N of the early warnings0And if the actual operation value of the stator winding temperature measuring point in the interval contained by the judgment counter is in a degradation trend, sending out an early warning signal, setting the abnormality calculator and the judgment counter to be zero, moving backwards for one moment, skipping to the step S2, and starting the next round of judgment.
In step S1, identifying the coefficient vector required for the on-line calculation of the stator winding temperature specifically refers to selecting a total n value of the coefficient vector, selecting continuous operation data of the stator winding temperature measurement points for a period of time, setting an iteration initial value of the coefficient vector, identifying the coefficient vector by a least square method or a neural network, judging whether a ratio of the number of errors smaller than a convergence error value to the total number reaches the standard after the calculation error is stable in the statistical identification process, if not, reselecting the total n value of the coefficient vector, and adjusting the parameter to continue the identification; if yes, the coefficient vector is output.
Running predicted value theta 'of stator winding temperature measuring point at moment t'tActual operation value theta of stator winding temperature measuring point at t momenttThe deviation of (2) is absolute deviation sigma, and the absolute deviation sigma is calculated by formula 4;
σ=|θt-θ'tequation 4
Where σ is the absolute deviation, θtIs an actual operation value theta of a stator winding temperature measuring point at the time t'tAnd (5) running a predicted value for the temperature measuring point of the stator winding at the time t.
Running predicted value theta 'of stator winding temperature measuring point at moment t'tActual operation value theta of stator winding temperature measuring point at t momenttThe deviation of (a) is calculated by using a relative deviation lambda which is calculated by formula 5;
λ=|θt-θ't|/θtformula 5
Where λ is the relative deviation, θtIs an actual operation value theta of a stator winding temperature measuring point at the time t'tAnd (5) running a predicted value for the temperature measuring point of the stator winding at the time t.
The embodiment is a further preferred embodiment, is suitable for different load working conditions of the flexible operation of the generator, and can be used for respectively modeling each temperature measuring point of the stator winding and extracting coefficient vectors which accord with respective operation characteristics, so that a high-precision calculation model is constructed, and the calculation precision is ensured.
Example 4
Referring to fig. 1, a stator winding temperature early warning method includes the following steps:
s1, setting an identification convergence error value and a ratio value, identifying a coefficient vector required by the on-line calculation of the stator winding temperature, and setting a threshold value K, a judgment interval total number N and an early warning total number N0;
S2, extracting n stator winding temperature continuous operation values, and calculating a stator winding temperature measuring point operation predicted value theta 'at t moment according to the coefficient vector through formula 1't;
Wherein, theta'tFor the stator winding temperature measuring point operation predicted value at the time t, thetat-1,θt-2…θt-nFor measuring actual running value, alpha, of stator winding temperature before time t1,α2…αnIn order to weight the vector of coefficients,as disturbance quantity, αiAs a vector of weighting coefficients, thetat-iThe actual operation value of the stator winding temperature measuring point at the time t-i is obtained;
s3, if the operation predicted value theta of the stator winding temperature measuring point at the moment t'tActual operation value theta of stator winding temperature measuring point at t momenttWhen the deviation is within the threshold value K, the actual operation value theta of the stator winding temperature measuring point at the time t is judgedtAnd (4) if the counter is judged to be added by 1 and the abnormal counter is not changed, the operation predicted value theta 'of the stator winding temperature measuring point at the moment t + 1't+1The calculation is determined by formula 2;
wherein, theta't+1The predicted value alpha of the operation of the stator winding temperature measuring point at the moment t +1iAs a vector of weighting coefficients, thetat+1-iIs the actual operation value of the stator winding temperature measuring point at the time t +1-i,is the disturbance quantity;
s4, if the operation predicted value theta of the stator winding temperature measuring point at the moment t'tActual operation value theta of stator winding temperature measuring point at t momenttWhen the deviation exceeds a threshold value K, the actual operation value theta of the stator winding temperature measuring point at the moment t is judgedtIf the abnormal data does not meet the normal characteristic, the abnormal data is judged, the counter is increased by 1, the abnormal counter is increased by 1, and the actual operation value theta of the stator winding temperature measuring point at the time t is eliminatedtAnd if the stator winding temperature measuring point does not participate in subsequent calculation, the operation predicted value theta 'of the stator winding temperature measuring point at the moment t + 1't+1The calculation is determined by formula 3;
wherein, theta't+1The predicted value alpha of the operation of the stator winding temperature measuring point at the moment t +1iAs a vector of weighting coefficients, thetat-iIs the actual operation value of the stator winding temperature measuring point at the time t-i,is the disturbance quantity;
s5, if the value of the abnormal counter is less than the total number N of the early warnings0And the value of the judgment counter is smaller than the total number N of the judgment intervals, and the step is shifted to S2 after moving a moment;
s6, if the value of the abnormal counter is less than the total number N of the early warnings0And if the value of the judgment counter is larger than or equal to the total number N of the judgment intervals, the actual operation values of the stator winding temperature measuring points of the total number N of the judgment intervals are considered to be normal, the abnormality calculator and the judgment counter are all set to zero, the operation is moved backwards by one moment, the step S2 is skipped, and the operation is startedJudging in one round;
s7, if the value of the abnormal counter is larger than or equal to the total number N of the early warnings0And if the actual operation value of the stator winding temperature measuring point in the interval contained by the judgment counter is in a degradation trend, sending out an early warning signal, setting the abnormality calculator and the judgment counter to be zero, moving backwards for one moment, skipping to the step S2, and starting the next round of judgment.
In step S1, identifying the coefficient vector required for the on-line calculation of the stator winding temperature specifically refers to selecting a total n value of the coefficient vector, selecting continuous operation data of the stator winding temperature measurement points for a period of time, setting an iteration initial value of the coefficient vector, identifying the coefficient vector by a least square method or a neural network, judging whether a ratio of the number of errors smaller than a convergence error value to the total number reaches the standard after the calculation error is stable in the statistical identification process, if not, reselecting the total n value of the coefficient vector, and adjusting the parameter to continue the identification; if yes, the coefficient vector is output.
Running predicted value theta 'of stator winding temperature measuring point at moment t'tActual operation value theta of stator winding temperature measuring point at t momenttThe deviation of (2) is absolute deviation sigma, and the absolute deviation sigma is calculated by formula 4;
σ=|θt-θ'tequation 4
Where σ is the absolute deviation, θtIs an actual operation value theta of a stator winding temperature measuring point at the time t'tAnd (5) running a predicted value for the temperature measuring point of the stator winding at the time t.
Running predicted value theta 'of stator winding temperature measuring point at moment t'tActual operation value theta of stator winding temperature measuring point at t momenttThe deviation of (a) is calculated by using a relative deviation lambda which is calculated by formula 5;
λ=|θt-θ't|/θtformula 5
Where λ is the relative deviation, θtIs an actual operation value theta of a stator winding temperature measuring point at the time t'tAnd (5) running a predicted value for the temperature measuring point of the stator winding at the time t.
The stator winding temperature at the time of the threshold value K and tMeasuring point actual operation value thetatThe actual operation value theta of the stator winding temperature measuring point at the time t in inverse proportiontThe larger the threshold K, the smaller the threshold K.
The total number of early warnings N0Less than or equal to the total number of judgment intervals N.
The embodiment is an optimal implementation mode, normal operation values of temperature at the previous moment are introduced, high-frequency signals do not need to be injected, the safe operation risk of the generator is avoided, the method is suitable for different load working conditions, the adaptability to each temperature measuring point of the stator winding is high, the individualized operation characteristics of each measuring point can be met, the local temperature of the stator winding is obtained through real-time online calculation, the calculation precision is high, and the effect of evaluating the health of the generator is greatly improved.
Claims (6)
1. A stator winding temperature early warning method is characterized by comprising the following steps:
s1, setting an identification convergence error value and a ratio value, identifying a coefficient vector required by the on-line calculation of the stator winding temperature, and setting a threshold value K, a judgment interval total number N and an early warning total number N0;
S2, extracting n stator winding temperature continuous operation values, and calculating a stator winding temperature measuring point operation predicted value theta 'at t moment according to the coefficient vector through formula 1't;
Wherein, theta'tFor the stator winding temperature measuring point operation predicted value at the time t, thetat-1,θt-2…θt-nFor measuring actual running value, alpha, of stator winding temperature before time t1,α2…αnIn order to weight the vector of coefficients,as disturbance quantity, αiAs a vector of weighting coefficients, thetat-iThe actual operation value of the stator winding temperature measuring point at the time t-i is obtained;
s3, if the operation predicted value theta of the stator winding temperature measuring point at the moment t'tActual operation value theta of stator winding temperature measuring point at t momenttWhen the deviation is within the threshold value K, the actual operation value theta of the stator winding temperature measuring point at the time t is judgedtAnd (4) if the counter is judged to be added by 1 and the abnormal counter is not changed, the operation predicted value theta 'of the stator winding temperature measuring point at the moment t + 1't+1The calculation is determined by formula 2;
wherein, theta't+1The predicted value alpha of the operation of the stator winding temperature measuring point at the moment t +1iAs a vector of weighting coefficients, thetat+1-iIs the actual operation value of the stator winding temperature measuring point at the time t +1-i,is the disturbance quantity;
s4, if the operation predicted value theta of the stator winding temperature measuring point at the moment t'tActual operation value theta of stator winding temperature measuring point at t momenttWhen the deviation exceeds a threshold value K, the actual operation value theta of the stator winding temperature measuring point at the moment t is judgedtIf the abnormal data does not meet the normal characteristic, the abnormal data is judged, the counter is increased by 1, the abnormal counter is increased by 1, and the actual operation value theta of the stator winding temperature measuring point at the time t is eliminatedtAnd if the stator winding temperature measuring point does not participate in subsequent calculation, the operation predicted value theta 'of the stator winding temperature measuring point at the moment t + 1't+1The calculation is determined by formula 3;
wherein, theta't+1The predicted value alpha of the operation of the stator winding temperature measuring point at the moment t +1iAs a vector of weighting coefficients, thetat-iIs the actual operation value of the stator winding temperature measuring point at the time t-i,is the disturbance quantity;
s5, if the value of the abnormal counter is less than the total number N of the early warnings0And the value of the judgment counter is smaller than the total number N of the judgment intervals, and the step is shifted to S2 after moving a moment;
s6, if the value of the abnormal counter is less than the total number N of the early warnings0If the value of the judgment counter is larger than or equal to the total number N of the judgment intervals, the actual operation values of the stator winding temperature measuring points of the total number N of the judgment intervals are considered to be normal, the abnormality calculator and the judgment counter are set to zero, the operation is moved backwards by one moment, the step S2 is skipped, and the next round of judgment is started;
s7, if the value of the abnormal counter is larger than or equal to the total number N of the early warnings0And if the actual operation value of the stator winding temperature measuring point in the interval contained by the judgment counter is in a degradation trend, sending out an early warning signal, setting the abnormality calculator and the judgment counter to be zero, moving backwards for one moment, skipping to the step S2, and starting the next round of judgment.
2. The stator winding temperature early warning method according to claim 1, characterized in that: in step S1, identifying the coefficient vector required for the on-line calculation of the stator winding temperature specifically refers to selecting a total n value of the coefficient vector, selecting continuous operation data of the stator winding temperature measurement points for a period of time, setting an iteration initial value of the coefficient vector, identifying the coefficient vector by a least square method or a neural network, judging whether a ratio of the number of errors smaller than a convergence error value to the total number reaches the standard after the calculation error is stable in the statistical identification process, if not, reselecting the total n value of the coefficient vector, and adjusting the parameter to continue the identification; if yes, the coefficient vector is output.
3. The stator winding temperature early warning method according to claim 1, characterized in that: running predicted value theta 'of stator winding temperature measuring point at moment t'tActual operation value theta of stator winding temperature measuring point at t momenttDeviation of (1) is adoptedAn absolute deviation σ calculated by equation 4;
σ=|θt-θ’tequation 4
Where σ is the absolute deviation, θtIs an actual operation value theta of a stator winding temperature measuring point at the time t'tAnd (5) running a predicted value for the temperature measuring point of the stator winding at the time t.
4. The stator winding temperature early warning method according to claim 1, characterized in that: running predicted value theta 'of stator winding temperature measuring point at moment t'tActual operation value theta of stator winding temperature measuring point at t momenttThe deviation of (a) is calculated by using a relative deviation lambda which is calculated by formula 5;
λ=|θt-θ’t|/θtformula 5
Where λ is the relative deviation, θtIs an actual operation value theta of a stator winding temperature measuring point at the time t'tAnd (5) running a predicted value for the temperature measuring point of the stator winding at the time t.
5. The stator winding temperature early warning method according to claim 1, characterized in that: the actual operation value theta of the stator winding temperature measuring point at the time of the threshold value K and the time ttThe actual operation value theta of the stator winding temperature measuring point at the time t in inverse proportiontThe larger the threshold K, the smaller the threshold K.
6. The stator winding temperature early warning method according to claim 1, characterized in that: the total number of early warnings N0Less than or equal to the total number of judgment intervals N.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110865263.9A CN113588123B (en) | 2021-07-29 | 2021-07-29 | Stator winding temperature early warning method |
PCT/CN2022/098498 WO2023005467A1 (en) | 2021-07-29 | 2022-06-14 | Temperature early warning method for stator winding |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110865263.9A CN113588123B (en) | 2021-07-29 | 2021-07-29 | Stator winding temperature early warning method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113588123A true CN113588123A (en) | 2021-11-02 |
CN113588123B CN113588123B (en) | 2023-01-06 |
Family
ID=78252026
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110865263.9A Active CN113588123B (en) | 2021-07-29 | 2021-07-29 | Stator winding temperature early warning method |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN113588123B (en) |
WO (1) | WO2023005467A1 (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114199505A (en) * | 2021-11-29 | 2022-03-18 | 中电华创(苏州)电力技术研究有限公司 | Generator stator bar circulation evaluation method based on correlation analysis |
WO2023005467A1 (en) * | 2021-07-29 | 2023-02-02 | 东方电气集团东方电机有限公司 | Temperature early warning method for stator winding |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108897954A (en) * | 2018-06-29 | 2018-11-27 | 龙源(北京)风电工程技术有限公司 | Wind turbines temperature pre-warning method and its system based on BootStrap confidence calculations |
CN111016946A (en) * | 2019-12-25 | 2020-04-17 | 新誉轨道交通科技有限公司 | Rail vehicle and temperature sensor fault detection system for air conditioner of rail vehicle |
CN112504511A (en) * | 2020-12-15 | 2021-03-16 | 润电能源科学技术有限公司 | Generator stator temperature monitoring method, device and medium |
CN112711832A (en) * | 2020-12-08 | 2021-04-27 | 重庆理工大学 | Method and system for early warning of temperature and fault identification of stator winding of synchronous generator |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP3979526B2 (en) * | 2002-06-19 | 2007-09-19 | 株式会社日本総合研究所 | Rotating machine thermal analysis method, rotating machine thermal analysis apparatus, computer program, and recording medium |
CN110414155B (en) * | 2019-07-31 | 2022-09-30 | 北京天泽智云科技有限公司 | Fan component temperature abnormity detection and alarm method with single measuring point |
CN113588123B (en) * | 2021-07-29 | 2023-01-06 | 东方电气集团东方电机有限公司 | Stator winding temperature early warning method |
-
2021
- 2021-07-29 CN CN202110865263.9A patent/CN113588123B/en active Active
-
2022
- 2022-06-14 WO PCT/CN2022/098498 patent/WO2023005467A1/en unknown
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108897954A (en) * | 2018-06-29 | 2018-11-27 | 龙源(北京)风电工程技术有限公司 | Wind turbines temperature pre-warning method and its system based on BootStrap confidence calculations |
CN111016946A (en) * | 2019-12-25 | 2020-04-17 | 新誉轨道交通科技有限公司 | Rail vehicle and temperature sensor fault detection system for air conditioner of rail vehicle |
CN112711832A (en) * | 2020-12-08 | 2021-04-27 | 重庆理工大学 | Method and system for early warning of temperature and fault identification of stator winding of synchronous generator |
CN112504511A (en) * | 2020-12-15 | 2021-03-16 | 润电能源科学技术有限公司 | Generator stator temperature monitoring method, device and medium |
Non-Patent Citations (1)
Title |
---|
李俊卿等: "同步发电机定子故障预警模型", 《电力科学与工程》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2023005467A1 (en) * | 2021-07-29 | 2023-02-02 | 东方电气集团东方电机有限公司 | Temperature early warning method for stator winding |
CN114199505A (en) * | 2021-11-29 | 2022-03-18 | 中电华创(苏州)电力技术研究有限公司 | Generator stator bar circulation evaluation method based on correlation analysis |
CN114199505B (en) * | 2021-11-29 | 2024-04-09 | 中电华创(苏州)电力技术研究有限公司 | Generator stator bar flow performance evaluation method based on correlation analysis |
Also Published As
Publication number | Publication date |
---|---|
WO2023005467A1 (en) | 2023-02-02 |
CN113588123B (en) | 2023-01-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113588123B (en) | Stator winding temperature early warning method | |
CN104537487A (en) | Assessment method of operating dynamic risk of electric transmission and transformation equipment | |
CN101447048A (en) | Method for predicting life of transformer insulation and management system thereof | |
CN112711832B (en) | Method and system for early warning of temperature and fault identification of stator winding of synchronous generator | |
CN114740303B (en) | Fault monitoring system of wireless passive high-voltage switch cabinet | |
CN107654342A (en) | A kind of abnormal detection method of Wind turbines power for considering turbulent flow | |
CN114050293B (en) | Working condition identification method of solid oxide fuel cell system | |
CN115293372A (en) | Photovoltaic string fault diagnosis method based on multi-dimension and multi-parameter numerical analysis | |
CN113156247B (en) | Early warning method and device for low-frequency oscillation of power system | |
CN110749810A (en) | Insulation fault prediction method and system for phase modulator | |
Cheng et al. | Multiblock dynamic slow feature analysis-based system monitoring for electrical drives of high-speed trains | |
CN106091515A (en) | The method that fired power generating unit once-though cooling circulation on-line operation optimizes | |
CN110750760B (en) | Abnormal theoretical line loss detection method based on situation awareness and control diagram | |
CN116338451A (en) | Steam turbine generator safety evaluation diagnosis system and method under deep peak regulation working condition | |
CN115932502A (en) | Method for evaluating insulation state of epoxy resin wall-penetrating sleeve in switch cabinet in damp and hot environment | |
CN113418632B (en) | Concept drift detection method for oil temperature prediction of oil immersed transformer | |
CN113591029A (en) | Stator winding temperature on-line calculation method | |
CN115796840A (en) | Green-energy thermoelectric equipment management platform based on data analysis | |
CN113344428B (en) | Health degree evaluation method for heat dissipation system of IGBT power module of wind power converter | |
CN113064075B (en) | Motor service life prediction method based on edge calculation and deep learning | |
Zhu | A new methodology of analytical formula deduction and sensitivity analysis of EENS in bulk power system reliability assessment | |
CN110136914A (en) | A kind of hybrid magnet interlock safety guard method | |
CN112084661A (en) | Wind turbine converter water cooling system cooling state assessment early warning method | |
CN112085336A (en) | Comprehensive capability assessment method and system for thermoelectric generator set in full-clean energy transformation process | |
Liu et al. | Evaluation of transformer state based on the life cycle |
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 |