Disclosure of Invention
The invention aims to provide a monitoring method and an early warning method for the temperature abnormity of a variable pitch motor, which aim to diagnose the abnormal working condition of a unit, discover hidden dangers as early as possible, reduce the fault frequency of the unit and improve the operation efficiency of the unit.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
a method for monitoring temperature abnormity of a variable pitch motor is characterized by comprising the following steps: calculating a temperature difference time sequence between the maximum average temperature value and the minimum average temperature value of each steady-state full-power segment in a certain time period in the wind turbine generator; and calculating the temperature difference average value of the temperature difference time sequence corresponding to each steady-state full-power segment, and searching temperature difference average value points exceeding the temperature difference average value fluctuation interval of the variable pitch motor according to the distribution condition of the temperature difference average values.
Further, the obtaining of the time series of temperature differences comprises: firstly, acquiring each steady-state full-power segment in a certain time period in a wind turbine generator, and then respectively calculating a temperature difference time sequence between the maximum average temperature value and the minimum average temperature value of each steady-state full-power segment; or, firstly, calculating an integral temperature difference time sequence between the maximum average temperature value and the minimum variable-pitch motor in a certain time period in the wind turbine generator, and then selecting the corresponding temperature difference time sequence under each steady-state full-power segment in the time period from the integral temperature difference time sequence.
Further, the obtaining of the steady-state full power segment comprises: the method comprises the steps of obtaining an actual operation power time sequence in a certain time period in a wind turbine generator, giving a lower limit of steady-state full-power generation, and selecting a segment with the output power of the generator set above the lower limit of the steady-state full-power generation and the duration time of the segment greater than half an hour as a steady-state full-power generation segment.
Further, the segments of the wind turbine generator in the initial stage of steady-state full power generation are removed from all steady-state full power generation segments.
Further, only the steady-state full-power segment of which the pitch motor temperature of the wind turbine generator is maintained within a certain temperature range is extracted from all the steady-state full-power segments.
Further, the selection of the temperature range comprises: and drawing a frequency distribution histogram of the temperature difference time sequence corresponding to each steady-state full-power segment, and selecting the temperature difference value corresponding to the group with the maximum density or the second largest density as the lower limit of the temperature range.
Further, the searching for the temperature difference average value point exceeding the temperature difference average value fluctuation interval of the variable pitch motor according to the distribution condition of the temperature difference average value comprises: calculating the average temperature difference value and the average wind speed value of the temperature difference time sequence corresponding to each steady-state full-power fragment, grouping the average temperature difference values corresponding to each steady-state full-power fragment according to the wind speed segments, drawing a scatter diagram of the average wind speed value and the average temperature difference value of each group of wind speed segments, and calculating the fluctuation interval of the average temperature difference value under each group of wind speed segments;
1/4 (q) calculating temperature difference average value data1)、1/2(q2)、3/4(q3) Quantile, interquartile range IQR ═ q3-q1And then, the fluctuation interval:
upper limit: uplim ═ q3+1.5IQR
Lower limit: q ═ q of Downlim1-1.5IQR
And searching for scattered points beyond the range of the fluctuation interval.
A pitch motor temperature abnormity early warning method is characterized by comprising the following steps: obtaining temperature difference time sequences corresponding to all steady-state full-power fragments from the variable pitch motor temperature anomaly monitoring method, and respectively calculating temperature difference average values of the corresponding temperature difference time sequences; and predicting the time when the future stable state of the variable pitch motor of the unit occurs and the mean value of temperature difference change by adopting a linear least square model and an ARIMA model respectively, and early warning according to the prediction.
Further, the mean value of the temperature difference of the variable pitch motor is predicted: respectively calculating temperature difference average values of corresponding temperature difference time sequences, combining the temperature difference average values according to the time sequences, and predicting by adopting an ARIMA model to obtain a predicted value of the temperature difference average value of the variable pitch motor at a future moment of the variable pitch system in a stable state and a confidence interval of the temperature difference average value of the variable pitch motor under a 95% confidence level;
predicting the occurrence time of the temperature difference stable state of the variable pitch motor: recording the occurrence time of each stable state in all temperature difference time sequences, and constructing a time interval sequence; predicting the future time of the time interval sequence by adopting a linear least square method to obtain the stable state occurrence time of the future time;
the early warning comprises the following steps:
if the predicted value of the future T moment is larger than a given threshold value, giving out early warning information of a variable pitch motor of the unit, otherwise, normally;
interval of confidence of predicted value if T time in future2,m=[am,bm]Corresponding to the actual normal operating temperature range [ c, d ] of the variable-pitch motor of the unit]And predicting the future temperature range of the variable pitch motor as follows:
Temp=[c+max(bm),d+max(bm)]
and if the Temp upper limit temperature reaches the fault alarm temperature of the unit, giving out early warning information of the variable pitch motor of the unit, otherwise, the unit is normal.
The invention has the beneficial effects that: according to the working principle of the variable pitch system, the temperature of the three variable pitch motors of the unit is taken as a research object, and the running conditions of the variable pitch system of the unit under different working conditions are deeply researched. The method is used for analyzing the difference of the operating temperatures of the three variable pitch motors of the unit and mining the change rule of the temperature difference time sequence of the variable pitch motors of the unit. Extracting variable pitch motor temperature difference data of a unit variable pitch system in a stable state, and researching: (1) the running state of the variable pitch system is greatly influenced by the wind speed, and the temperature difference of three variable pitch motors of the unit changes along with the change rule of the wind speed; (2) the trend that the temperature difference of three variable pitch motors of the unit changes along with time is researched, and the purpose is to diagnose abnormal working conditions of the unit, discover hidden dangers as early as possible, reduce the fault frequency of the unit and improve the operation efficiency of the unit. The relation between the wind speed and the temperature difference researched by the method can diagnose the performance of the variable pitch system of the unit under different working conditions in the operation process, and the operation state of the variable pitch system of the unit under each working condition can be monitored in a targeted manner; in addition, the method can find the operation difference of the three variable pitch motors in the operation process of the unit, better show the operation state of the variable pitch system of the unit and discover the potential safety hazard of the variable pitch system of the unit earlier. According to the method, the stable state of the temperature change of the variable pitch motor is searched according to the full-power segment of the power; under the stable state of the unit variable pitch motor, the change condition of the temperature difference of the variable pitch motor and the performance state of the temperature difference under different wind speed states are deeply researched. The method combines the actual performance of the units, evaluates the running state of the steady-state variable pitch motor under different wind speed states of each unit through screening of outliers, and quickly searches the abnormal working condition of the units; and then, predicting the time when the future stable state of the variable pitch motor of the unit occurs and the mean value of temperature difference change by adopting a linear least square model and an ARIMA model, and diagnosing the state of the variable pitch motor of the unit.
Detailed Description
For a better understanding of the present invention, the technical solutions of the present invention will be further described below with reference to the following examples and accompanying drawings.
Example one
A monitoring method and an early warning method for temperature abnormity of a variable pitch motor are shown in figure 1.
The relation between the wind speed and the temperature difference researched by the method can diagnose the performance of the variable pitch system of the unit under different working conditions in the operation process, and the operation state of the variable pitch system of the unit under each working condition can be monitored in a targeted manner; in addition, the method can find the operation difference of the three variable pitch motors in the operation process of the unit, better show the operation state of the variable pitch system of the unit and discover the potential safety hazard of the variable pitch system of the unit earlier.
Structure of temp. difference time sequence
For a single wind generating set, the operation conditions of the three blades are basically consistent. In the SCADA fault alarm system, the monitoring is specially carried out aiming at the asymmetry of the variable pitch angle and the variable pitch speed of the three blades, and a fixed fault grade is set. The operation state of the variable pitch motor is monitored only by aiming at the maximum temperature of the variable pitch motor, alarm is given according to the maximum temperature of the variable pitch motor of the unit, and a lot of important information is inevitably lost if the operation temperatures of the three variable pitch motors are consistent and no operation is carried out.
According to the method, on the aspect of index selection reflecting the running state of a variable-pitch motor, the running performance of the variable-pitch motor of the unit is fully displayed according to the running state of a variable-pitch system of the unit, and two temperature indexes of the variable-pitch motor with higher temperature and lower temperature are selected for analysis and recorded as follows:
(1) the temperature time sequence of the variable pitch motor with the highest average value of the temperature data of the variable pitch motor of the unit is as follows:
{X}={x1,x2,...,xt};
(2) the temperature time sequence of the variable pitch motor with the lowest average value of the temperature data of the variable pitch motor of the unit is as follows:
{Y}={y1,y2,...,yt};
(3) time series of temperature differences:
{E}={e1,e2,...,et}={x1-y1,x2-y2,...,xt-yt}。
second, data selection under steady state full power state
The variable pitch system mainly comprises the following functions: the wind power generation set is ensured to effectively utilize wind energy, the safety of the fan under extreme conditions is ensured, and overspeed runaway and tower collapse accidents are avoided. When the wind speed reaches the rated wind speed, the unit starts the variable pitch system, so that the output power of the unit is controlled to be close to the rated power, and the whole operation state of the unit is stable. Therefore, the method has practical significance in researching the running state of the variable pitch system near the rated power.
The driving of the variable pitch system mainly depends on a variable pitch motor, and the temperature can be increased in the running process of the motor. When the variable pitch system operates for a period of time, the heat emitted by the heat dissipation system is constant, the temperature of the motor is not increased any more, and the variable pitch system can stably operate near a certain temperature.
Based on the principle, the method selects the stable operation state of the variable pitch system of the unit (namely the stable state of the variable pitch system), and researches the change condition of the temperature difference of the variable pitch motor.
Data were screened as follows:
(1) it is generally considered that the steady full-power stage of the unit is when the output power of the unit is stabilized near the rated power for more than a few and a half hours. According to the actual operation Power time sequence Power, a lower limit p of steady-state full Power is givenlimSelecting a segment P (P) with the output power of the unit above the lower limit of the steady-state full power and the duration exceeding T (T is more than 0.5h)1,P2,...,Pn}。
(2) Selecting a variable pitch motor temperature difference sequence corresponding to the steady-state full-power segment in the step (1) to obtain a set variable pitch motor temperature difference data segment combination { Z } ═ Z in the steady-state full-power state1,z2,...,zn}。
(3) Considering that the temperature of the variable pitch motor can be maintained to be operated near a certain temperature after the unit is operated for a period of time, for the sequence { Z } in (2), a segment of the unit in a certain period of time just starting from the steady-state full-time start (namely before the temperature is stabilized) needs to be removed or only a time segment of the variable pitch motor temperature of the unit which is maintained to be fluctuated near a certain temperature is extracted. The time segment that the temperature of the variable pitch motor of the unit is maintained to fluctuate near a certain temperature is extracted, the distribution of the segment is obvious, the numerical value fluctuates in a certain range, and the temperature difference segment of the variable pitch motor after a period of steady-state operation is convenient to find:
selecting the ith running segment Z in the { Z }i={ei,1,ei,2,...,ei,t-wherein i ═ 1, 2.., n, frequency distribution histograms are plotted;
determining the number of groups according to an empirical formula provided by Sturges, drawing a frequency distribution histogram, and selectingOne of the first two groups with higher density corresponds to the lowest temperature, i.e. the lower temperature limit Qlim,i;
③ removing Qlim,iThe point that Q or more is retainedlimData point Q ofiThe data points thus screened are relatively concentrated. Through the operation, the variable pitch motor temperature difference segment Q ═ Q after steady-state operation for a period of time is obtained1,Q2,...,Qn}
Temperature anomaly monitoring of variable pitch motor
The running states of the variable pitch system of the unit have certain differences in the performance of different wind speed sections, the running differences of the unit in different wind speed sections are further searched, hidden dangers of the variable pitch system can be found as soon as possible, operation and maintenance personnel can conveniently and pertinently overhaul the unit, and overhaul time is shortened. When the unit is in a steady-state full-firing stage, the influence of the wind speed on a variable pitch system of the unit is large, so that the research on the relation between the wind speed and the temperature difference of a variable pitch motor is of practical significance.
Calculating the average temperature difference value and the average wind speed value of the temperature difference time sequence corresponding to each steady-state full-power fragment, grouping the average temperature difference values corresponding to each steady-state full-power fragment according to the wind speed segments, drawing a scatter diagram of the average wind speed value and the average temperature difference value of each group of wind speed segments, and calculating the fluctuation interval of the average temperature difference value under each group of wind speed segments; because the selected data are segments of the unit variable pitch system which operate stably, the temperature difference change of the unit should be kept within a certain fluctuation range. Through data representation, it is possible to obtain:
(1) in a stable state of a unit variable pitch system, the fluctuation Interval of the mean value of the temperature difference of variable pitch motors of different wind speed sections1,
(2) According to the distribution in the data, if the state θ is { W, E }, where W represents a wind speed section and E represents a temperature difference, and if the data deviates from the distribution, the operation of the unit pitch motor in the wind speed state is abnormal, and therefore attention needs to be paid.
And (3) determining the scatter point: 1/4 (q) calculating temperature difference average value data1)、1/2(q2)、3/4(q3) Quantile, interquartile range IQR ═ q3-q1And then, the fluctuation interval:
upper limit: uplim ═ q3+1.5IQR
Lower limit: down lim q1-1.5IQR
And searching for scattered points beyond the range of the fluctuation interval.
Fourth, early warning of abnormal temperature of variable pitch motor
If the fluctuation range of the temperature difference of the variable pitch motor is too large in the stable state of the variable pitch system of the unit, when the external environment temperature rises, the temperature of the variable pitch motor of the unit can easily reach the alarm value of the SCADA fault system of the unit, and the unit is prompted to report faults and even shut down. The fluctuation of the temperature difference of the variable pitch motor can be measured by using a fluctuation interval of the temperature difference; and then the mean value and the occurrence time of the temperature difference change of the variable-pitch motor of the unit in the future stable state are respectively predicted, so that the hidden danger of the variable-pitch system of the unit can be discovered earlier.
Obtaining each section Q of the temperature difference of the variable pitch motoriAnd respectively calculating the average values of the corresponding indexes, and then combining according to the time sequence. Considering that the change factors of the temperature difference average value influencing the stable state of the variable pitch system of the unit also include time factors, the temperature difference sequence has the self-correlation characteristic, so that:
(1) mean value prediction of temperature difference of variable pitch motor
Predicting by adopting a time series moving average autoregressive model (ARIMA model) to obtain future T of the average value of the temperature difference of the variable pitch motor of the variable pitch system of the unit under the stable statem(m is 1,2,..) the predicted value of the moment, and a confidence Interval under the confidence level of 95% of the mean value of the temperature difference of the variable-pitch motor of the unit1。
(2) Prediction of variable pitch motor temperature difference stable state occurrence time
Recording the moment of occurrence of each stable state in the variable pitch motor temperature difference segment Q as sigma ═ sigma1,σ2,...,σnAnd (4) constructing a time interval sequence:
Δ={Δ1,Δ2,...,Δn}={σ1-σ1,σ2-σ1,σ3-σ1,σ4-σ1,...,σn-σ1}
applying linear least square method to future T of delta sequencem(m 1, 2..) the time is predicted. By prediction, the future T is knownmThe steady state occurrence time at the moment is sigman+j=Δn+j+σ1(j=1,2,...,m)。
(3) Judging (early warning) the indexes for evaluating the temperature difference:
if the predicted value of the future T moment is larger than a given threshold value
The variable pitch motor of the unit needs to pay attention, otherwise, the variable pitch motor is normal;
secondly, if the confidence Interval of the predicted value of the future T moment of the temperature difference of the variable pitch motor2,m=[am,bm]Wider, corresponding to the actual normal operating temperature range [ c, d ] of the variable pitch motor of the unit]And predicting the possible future temperature range of the variable pitch motor as follows:
Temp=[c+max(bm),d+max(bm)]
if the Temp upper limit temperature is close to the temperature of the unit fault alarm, the variable pitch motor of the unit needs to pay attention, otherwise, the variable pitch motor is normal.
Example two
Embodiment combines variable pitch motor temperature abnormity monitoring method and early warning method of certain number 07 wind turbine generator example
(1) Selecting SCADA data of a certain wind field, wherein the machine type is 1.5MW of golden wind, the time interval is 10s, and the time interval is from 2018-05-1900:00: 00-2019-04-3023: 59: 50. The running state of the unit in 5 months in 2018 is selected for displaying, the rated power of the unit is 1.5MW, and the lower limit of the steady-state full-power generation power is set to be plim1450kw, the duration time exceeds T and is more than or equal to 1h, and the temperature of the variable pitch motor corresponding to the selected steady-state full power segment is selected;
(2) time sequence for calculating temperature difference between maximum and minimum variable-pitch motor temperature
In the operation process of the unit, the temperature of the variable pitch motor 1 is always higher than that of other variable pitch motors, so that the time sequence corresponding to the { X } is the temperature time sequence of the variable pitch motor 1, and the time sequence corresponding to the { Y } is the temperature time sequence of the variable pitch motor 3. And calculating to obtain a temperature difference sequence { E }.
(3) Each full hair segment ziThe temperature difference time sequence of the variable pitch motor is drawn into a histogram, and the numerical value of the second rank is selected as the lower limit value for screening the temperature difference of the variable pitch motor. As shown in FIG. 2, the lower temperature limit Q is selectedlim,i=45℃。
(4) According to the lower limit value Q of the temperature difference of the variable pitch motor in the step (3)lim,iAnd selecting numerical value segments with the temperature difference higher than the lower limit value in the full-transmission segments to obtain a sequence Q, and calculating the average value of the temperature difference and the average value of the wind speed of the variable pitch motor in each segment, as shown in figure 3.
(5) And grouping the sequence Q according to the wind speed, and calculating the average temperature difference value of the variable pitch motor. And drawing a scatter diagram of the average value of the wind speed and the average value of the temperature difference of the variable pitch motor. Interval of fluctuation Interval of variable pitch motor temperature difference of different wind speed sections1=[32.9,49.1]IQR 2.87, Uplim 46.5, and Downlim 35.1. As can be seen from FIG. 4, when the wind speed of the unit is between 16.5 m/s and 18.5m/s, the operating condition of the unit is abnormal, and important attention needs to be paid.
(6) Selecting T
mSelecting empirical data as 4
Significance levels were selected at 0.05 and 0.2.
Test shows that ARIMA (1,2,1) model is selected for modeling (as shown in FIG. 5), and the residual sequence of the model is verified to be white noise. Calculating to obtain Interval1The results are shown in the following table.
Secondly, modeling the stable state occurrence moment of the variable pitch motor by adopting a linear least square method (as shown in figure 6), wherein the linear model y is the R of ax + b2The regression coefficients were all significant at 0.98, and the results are shown in the table below.
Prediction result summary of mean temperature difference and stable state occurrence time of variable pitch motor
Thus, it can be seen that:
firstly, the predicted value of the temperature difference of the variable pitch motor of the future 4 steady-state full-power-generation sections is between 27.9 and 32.6 ℃, is greater than the given 15 ℃, and the temperature difference is large;
the normal operation temperature of a variable pitch motor of the unit is [ c, d ]]=[50,70]The confidence interval at which the 95% confidence level is selected [ a DEG C ]m,bm],max(bm) 53.1 ℃. And (3) integrating the temperature difference confidence interval and the normal operation temperature, wherein the maximum temperature Temp of three variable pitch motors is (103.1, 123.1) in 4 steady-state full-power-generation sections of the unit in 4 months in 2019]And the risk of exceeding the alarm value of the unit exists, and the variable pitch motor of the unit needs to pay attention.
The above description is only an application example of the present invention, and certainly, the present invention should not be limited by this application, and therefore, the present invention is still within the protection scope of the present invention by equivalent changes made in the claims of the present invention.