JP3019606B2 - Rotating machine temperature rise test method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for testing a temperature rise of a rotating electric machine such as an induction motor.

[0002]

2. Description of the Related Art Conventionally, a temperature rise test of a rotating electric machine has been performed based on a test method defined by standards such as an actual load method and an equivalent load method.

The relationship between the temperature rise and time in this case is almost as shown in FIG. 6, and the time required from the start of the test to the end of the test is almost 3 to 4 hours in most cases.
The main reason is that, depending on the standard, after the temperature rise approaches the final value, the change in temperature rise per hour Δθ is 1
This is because the temperature is required to be within ℃.

Next, a conventional temperature rise test method will be described using an induction motor as an example. The temperature rise test of the induction motor is performed with a circuit configuration as shown in FIG. In the figure, IM is the machine under test (induction motor), T is the machine under test IM
, A series insertion transformer for supplying a current equal to the rated current to the
Frequency f ₂ different from its original frequency f ₁
Generator for auxiliary power supply for superimposing current of C1, C2
Is the breaker, a is the connection point of the voltmeter / ammeter / wattmeter, f
₁ power supply frequency (nominal frequency), f ₂ is superimposed frequency, V ₁ is the terminal voltage (rated voltage), V ₂ is a voltage superimposed, I ₁ is the primary current supplied to the tester IM, W is input Power.

A voltage V ₁ corresponding to a rated voltage supplies an iron loss to the device under test IM connected in this way, and a copper loss is supplied by a primary current I ₁ corresponding to a rated current. Will be. With this, the machine under test IM
Is simulated in the same condition as when a rated load is applied, and a temperature rise test is performed under such conditions.

In a temperature rise test for a medium capacity machine or a large capacity machine, a temperature rise pattern as shown in FIG. 6 is usually exhibited, and three to four hours are required from the start of the test to the end of the test. During this time, power corresponding to the loss of the device under test IM is generated inside the device under test IM, and the driving efficiency is 9%.
From 0 to 96%, a loss of 4 to 10% of the power is consumed over 3 to 4 hours.

By the way, in the case of an output of 1000 kW, the efficiency is about 95%, and if the time required for the test is 4 hours, it is as large as 1000 × (1−0.95) × 4 = 200 (kWH). Power will be consumed.

[0008]

The conventional method for testing the temperature rise of a rotating electric machine has been performed by the above method.
The time required for the test was long, and the power consumption was large, so that the economic burden was quite severe.

For example, if the annual average motor output is 1000 kW (average efficiency is 94%) and the average annual production is 500 units, the power required for the total temperature rise test is approximately as follows.

[0010] 1000 x (1-0.94) x 4 x 500 = 120000 (kWH) Then, when the cost required for this power consumption is calculated, it becomes 20 yen / kWH x 120,000 kWH = 2.4 million yen. In addition, 2 per worker per hour per worker
Assuming a labor cost of 10,000 yen, a total labor cost of 20,000 yen x 500 x 4 = 40 million yen is required.
This results in extremely high test costs.

SUMMARY OF THE INVENTION The present invention has been made in order to solve the above problems, and an object of the present invention is to provide a method for testing a temperature rise of a rotating electric machine which can reduce the time required for the test and reduce the power consumption. With the goal.

[0012]

Means for Solving the Problems] Temperature rise test method of a rotating electric machine according to the present invention, the time differential d [theta] / dt at the time as well as measuring the temperature actual value theta _c from the temperature rise test started after a predetermined time Calculated and these measured temperature values θ _c
The required temporal change pattern of the thermal resistance R is selected from the time derivative dθ / dt, and the model calculated temperature θ after a predetermined time is calculated using the temporal change pattern of the thermal resistance R, and the predetermined time is reached. which was compared with the temperature measured value theta _c and the model calculated temperature theta in time, the deviation ε is characterized in that the final predicted temperature theta _m constructed as calculated output when within a predetermined range It is. Further, when the deviation ε is not within the predetermined range, another temporal change pattern of the thermal resistance R is selected and the same processing as described above is performed.

[0013]

According to the above configuration of the present invention, the temporal change pattern of the thermal resistance R, which is the largest factor of the error in the temperature rise test, is centered, and the model calculated temperature θ calculated based on the pattern is the measured temperature value θ. if substantially coincident is _c, it can be estimated with high accuracy at an early stage the final predicted temperature theta _m from the start of the test.

[0014]

DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below in detail with reference to the drawings.

FIG. 1 shows an outline of an apparatus used in carrying out a temperature rise test method for a rotating electric machine. In FIG.
1 is a stator core of the rotating electric machine to be measured, 2 is a stator coil housed in the stator core 1, and 3 is a temperature detection sensor (usually a thermocouple or A search coil), 4 is a lead wire derived from the temperature detection sensor 3, 5 is a stator coil temperature measurement device (fixed system) connected to the temperature detection sensor 3 via the lead wire 4, and A is the corresponding temperature rise. Test equipment.

The temperature rise test apparatus A includes a temperature rise calculation model 6 (variable system) that outputs a temperature θ calculated by calculation according to a predetermined model, an actual temperature measured value θ _c by the stator coil temperature measurement apparatus 5 and a model. Deviation ε from calculated temperature θ (= θ
_c− θ), adaptive algorithm 8, temperature rise test time measuring device 9, [dθ / dt] calculating device 10,
Temperature rise value calculation device 11, thermal resistance calculation device 12, temperature rise calculation parameter input / output device 13, thermal resistance storage device 1
4. It has a temperature rise test data storage device 15 and the like.

Next, the principle of the temperature rise test method for a rotating electric machine according to the present invention will be described. Here, the structure of the rotating electric machine will be described in a simplified manner.

P: Loss generated by rotating electric machine [kW] C _c : Specific heat of stator coil [J / kg ° C] _Wc : Mass of stator coil [kg] R: Thermal resistance of insulator [° C / W] C _i : Specific heat of the iron core [J / kg ° C.] W _i : Mass of the iron core [kg] Here, assuming that the model calculation temperature t seconds after the start of the test is θ, PΔt = (C _c W _c + C _i W _i ) When fix Δθ + (θ / R) Δt (J) it in the form of differential equations, when obtaining the _{P = (C c W c +} C i W i) dθ / dt + θ / R than this calculated temperature theta, theta = R {P− (C _c W _c + C _i W _i ) dθ / dt} (1) (1) In the _{equation, P, C c, W c} , C i, W
_i is a known value obtained before the test, and is a value specific to the rotating electric machine to be measured.

The main cause of the error in the equation for calculating the temperature rise is the thermal resistance R, and how to determine the thermal resistance R affects the calculation error of the model calculation temperature θ. By the way, when the temperature rise test is started, the calculated temperature θ changes every moment as the time t elapses. However, it is convenient that the temperature elapses after a relatively short time period (2 to 5 minutes) immediately after the test. , The calculated temperature θ is the thermal resistance R of the rotating electric machine to be measured.
It is known that the temperature rises along a curve according to the cooling capacity.

[0020] Temperature rise test in the rotating electric machine, wherein the _{P, C c, W c,} C i, since the number of rotation to be changed to W _i during the test is constant i.e. in the cooling air flow rate is constant, the temperature Found An important parameter for the deviation between θ _c and the calculated temperature θ based on the calculation of equation (1) is the thermal resistance R.

[0021] Therefore, while sequentially measuring the temperature reading theta _c, correcting the thermal resistance R of the deviation between the temperature measured value theta _c and calculating temperature theta obtained by calculation is as close as possible to zero (1) Then, the final predicted temperature θ _m can be obtained with high accuracy.

More specifically, in FIG.
_It is assumed that the measured temperature value when it becomes ₁ is θ _c1 . This based on the temperature measured value theta _c1 and a predetermined thermal resistance R calculated expected temperatures at time t ₂ to time units has elapsed, to the calculated temperature and theta _2. The deviation ε (=) between the actually measured temperature θ _c2 at the time t ₂ and the predicted calculated temperature θ ₂
θ _c2 -θ ₂₎ has, when the deviation ε exceeds a predetermined range, to correct the thermal resistance R of the parameters.
Then, the calculated temperature θ ₃ of the prediction at the next time t ₃ becomes more accurate.

Incidentally, as shown in FIG. 3, the thermal resistance R changes due to an increase in the internal temperature of the rotating electric machine over time. However, if the time elapses beyond a certain value, the internal temperature also equilibrates, and the thermal resistance R converges to a constant value. As shown by A, B, C, D,... In the figure, the temporal change pattern of the thermal resistance R differs depending on the type of the rotating electric machine, the cooling method, and the like. Then, if conditions such as a model and a cooling method are determined, it is possible to select a certain one as a temporal change pattern of the thermal resistance R.

By accumulating the temperature rise tests for various rotating electric machines, it is possible to accumulate various patterns of data as a temporal change pattern of the thermal resistance R.

The temporal change pattern of the thermal resistance R is formed into a mathematical expression and stored in the thermal resistance storage device 14. Then, in the actual temperature rise test, if the optimal temporal change pattern of the thermal resistance R is selected from the thermal resistance storage device 14 and applied, the model calculation temperature θ can be calculated with higher accuracy. . Further, as the number of temperature rise tests increases, the number of accumulated temporal change patterns of the thermal resistance R also increases, and the model calculation temperature θ can be calculated with higher accuracy.

In the temperature rise test, as shown in FIG. 2, a minute temperature rise dθ at a minute time dt from time t ₁ is measured. That is, dθ / dt is calculated. On the other hand, since the temperature at that time is measured as the measured temperature value θ _c , the equation (1) is expressed as follows: θ _c = R {P− (C _c W _c + C _i W _i ) _d θ / dt} R can be obtained by back calculation. Here, P, C _c , W _c , C _i , and W _i are assumed to be unchanged.
It can be determined thermal resistance R also in the subsequent time t _2. The thermal resistance R of the period from the time t ₁ to time t ₂
Of the thermal resistance R to be selected can be determined based on the change of the thermal resistance R. There is a case where, for example, the temporal change pattern C is selected up to a certain point in time, but the temporal change pattern D is selected from the next period. This is the parameter-"heat resistance R
Correction ".

Now, when the equation (1) is calculated to obtain the model calculated temperature θ, θ = RP [1−exp ＝ −t / R (C _c W _c + C _i W _i )}]... 2)

FIG. 2 is a graph of equation (2). Here, τ is a time constant, and τ = R (C _c W _c + C _i W _i ). Equation (2) includes two Rs. R is a function of time t. However, it is known that when the time τ corresponding to the time constant elapses from the start of the temperature rise test, the time change pattern of the thermal resistance R is determined. When the temporal change pattern of the thermal resistance R is temporarily determined (the determined R is defined as _R0 ), 3 to 4 hours, which is a time required for a general temperature rise test, is calculated by the equation (2). substituted in the t (this t and t _0), and calculates the final predicted temperature theta _m. That, θ _m = R ₀ · P _{[1- exp {-t 0 / R 0} (C c W c + C i W i)} ] ............ becomes (3), the final predicted temperature theta _m is determined I do. By the way, t
Θ _m when → ∞, exp (−∞) → 0, θ _m (∞) → R ₀ · P, and when t = τ, θ is exp (−1) −１1/2 .
Since 718 ≒ 0.37, θ _m (τ) = 0.63 · R ₀ · P = 0.63 · θ _m (∞).

[0029] The final predicted temperature theta _m is intended to be calculated when only the time tau constant has elapsed time from the start of the temperature rise test, when the rotary electric machine, typically, the time constant tau 30-4
Since the time required is about 0 minutes, the required time reduction ratio is (30-40) ÷ ｛(3-4) × 60｝ ≒ 1/6, that is, about 1/6 time And save a lot of time.

If an appropriate temporal change pattern of the thermal resistance R is determined before the time of the time constant τ elapses, the final predicted temperature θ _m with high accuracy can be determined even earlier. .

However, in the case of a rotating electric machine whose type and cooling method are completely new, a temporal change pattern of the thermal resistance R corresponding to the rotating electric machine is not accumulated. The temperature rise test will be continued until the change in rise falls within 1 ° C. However, this requires only one, new since the temporal change pattern of the thermal resistance R is registered, to confirm the temporal change pattern of the thermal resistance R early the next time, determine the final predicted temperature theta _m Will be able to do it.

The detailed operation will be described with reference to the flowchart of FIG.

The temperature rise test is started, and step S
1 after waiting for the predetermined time T1 at reading the temperature measured value theta _c proceeds to step S2. The time is measured by the temperature rise test time measuring device 9. The measured temperature value θ _c is taken in from the stator coil 2 via the stator coil temperature measuring device 5. In step S3, the [dθ / dt] arithmetic unit 10 calculates dθ / dt from the start to the time. Then, in step S4, the thermal resistance calculating device 12
, The value of the thermal resistance R is calculated from θ _c and dθ / dt, the temporal change pattern of the thermal resistance R is converted into a mathematical expression in step S5, and the past temperature rise test stored in the thermal resistance storage device 14 in step S6. Thermal resistance R calculated in the data
Of the thermal resistance R having the value of R closest to the value of R is selected. The selected temporal change pattern of the thermal resistance R is sent to the temperature rise value calculation device 11 via the temperature rise calculation parameter input / output device 13.

Then, in step S7, the temperature rise value calculation device 11 calculates a model calculation temperature θ for prediction after T minutes by using the selected temporal change pattern of the thermal resistance R. Send in. In after step S8 in T minutes compares the temperature reading theta _c and calculated temperature theta. The comparison is performed in the adder 7. That is, the deviation ε between the actual measured value θ _c taken from the stator coil temperature measuring device 5 and the calculated temperature θ taken from the temperature rise calculation model 6 is obtained. The adaptive algorithm 8 is
In step S9, the deviation ε between the measured temperature θ _c and the calculated temperature θ
Is determined to be within a predetermined range. Deviation ε
Is within the predetermined range, the previously selected thermal resistance R
Was confirmed to be correct over time,
It proceeds to step S10 by using a temporal change pattern of the selected thermal resistance R to calculate the final predicted temperature theta _m, determines whether to terminate the temperature rise test at step S14. These processes are performed in a loop of the temperature rise calculation model 6, the adder 7, and the adaptive algorithm 8. If it is determined that the temperature rise test has not been completed, the process returns to step S7. When good exit temperature rise test is to store the current temperature rise test data on temperature rise test data storage device 15 in step S15, the display device Ya temperature rise test data, including the final predicted temperature theta _m in step S16 Output to printer and end temperature rise test. In addition,
Upon output of the temperature rise test data may also be output by evaluating the correctness degree of final predicted temperature theta _m of temporal change pattern of the thermal resistance R was selected.

If the deviation .epsilon. Does not fall within the predetermined range in step S9, the flow advances to step S11 to determine whether or not another temporal change pattern of the thermal resistance R can be selected. If it is possible, the process returns to step S7 to repeat the same operation. If not possible, step S1
2, the temporal change pattern of the thermal resistance R in the test up to the time T1 is adopted, and in step S13, the thermal resistance R
Using a temporal change pattern to calculate the final predicted temperature theta _m, or less, the process proceeds to step S14.

In the above embodiment, the thermal resistance R is used as a parameter.
Only the temporal change pattern was considered. It is P, C
_c, W _c, C _i, such as W _i is because a factor determined uniformly if the test conditions of the measured rotary electric machine Sadamare. Usually, once set, there is no need to change it. However, event if there is miscalculation, only the selection of the temporal change pattern of the thermal resistance R can occur if the temperature measured value theta _c and the model calculated temperature theta not be calculated correctly. In such a case,
As parameters other than the thermal resistance R, P, C _c , W _c ,
Priorities may be assigned to C _i and W _i, and the next parameter to be corrected next to the thermal resistance R may be sequentially corrected. If the temporal change pattern of the thermal resistance R does not match the physical phenomenon, a warning may be issued to notify that parameters other than the thermal resistance R are abnormal.

[0037]

As described above, according to the present invention, the time required for the test can be greatly reduced, and the power consumption can be greatly reduced.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an outline of an apparatus used for a temperature rise test of a rotating electric machine (induction motor) according to one embodiment of the present invention.

FIG. 2 is a diagram illustrating a method of predicting a calculated temperature and a temperature rise characteristic of a model calculated temperature in the example.

FIG. 3 is a diagram showing a temporal change pattern of a thermal resistance R in an example.

FIG. 4 is a flowchart for explaining the operation in the embodiment.

FIG. 5 is a circuit diagram of a conventional induction motor in a temperature rise test.

FIG. 6 is a general temperature rise characteristic diagram in a temperature rise test of a rotating electric machine.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Stator core 2 Stator coil 3 Temperature detection sensor 4 Lead wire 5 Stator coil temperature measurement device 6 Temperature rise calculation model 7 Adder 8 Adaptive algorithm 9 Temperature rise test time measurement device 10 [dθ / dt] calculation device 11 Temperature Temperature rise calculation device 12 Thermal resistance calculation device 13 Temperature rise calculation parameter input / output device 14 Thermal resistance storage device 15 Temperature rise test data storage device A Temperature rise test device

Claims (1)

(57) [Claims] [Claim 1] to calculate the time derivative d [theta] / dt at the time as well as measuring the temperature rise test starting temperature measured value theta _c after a predetermined time from these temperature reading theta _c and time derivative d [theta] / dt A required temporal change pattern of the thermal resistance R is selected, a model calculated temperature θ after a predetermined time is calculated using the temporal change pattern of the thermal resistance R, and when the predetermined time is reached, the measured temperature value θ _c Is compared with the model calculated temperature θ, and when the deviation ε does not fall within the predetermined range, the temporal change pattern of the thermal resistance R is corrected until the deviation ε falls within the predetermined range, so that the deviation ε falls within the predetermined range. temperature rise test method of a rotating electric machine, characterized by being configured to final predicted temperature theta _m was calculated output when it becomes.