JPS58172167A - Controller for elevator - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field of the Invention] The present invention relates to an elevator control device that provides a predetermined speed and acceleration pattern.

[Technical background of the invention]

FIG. 1 shows the configuration of an example of a conventional elevator control system.

In the figure, reference numeral 1 denotes a reference speed pattern generator that generates a reference speed pattern signal for the elevator.

2 is a relay contact for switching between reference speed and turn and landing and turn, 3 is an addition/subtraction point between the reference signal which is the output of the reference speed and turn generator 1 and the feedback signal, and 4 is the point 3 of point 3. 5 is a thyristor Leonard control device that generates an armature voltage of the hoisting DC motor; 6 is a position detection device that detects the position of the car; and 7 is a governor. Reference numeral 8 represents a DC motor for the hoisting machine, which performs constant excitation control.

9 is an interpol winding for taking out the current anti-hunt signal;
11 is a counterweight, 10 is a sheave directly connected to the shaft of the hoist 8, 12 is a ladder, and 13 is a landing pattern generator that generates a reference pattern signal near the landing position based on the expected landing position. . 14 It is a load detector installed under the floor of the car 12, and the car floor has a floating structure.
Detects the displacement of the anti-vibration rubber inserted between the car frame and
The load inside the car 12 is detected. 15 is a voltage anti-hunt application circuit that detects the terminal voltage from a pilot brush provided on the commutator of the hoisting motor 8 and applies voltage negative feedback and voltage anti-hunt application; 16 is a voltage anti-hunt application circuit taken out from both ends of the interpol winding 9; 17 is a landing signal generator which synchronously rectifies the signal obtained by the landing start turn generator 13 and uses it as a reference signal at the time of landing. , 18 is a load/thousand turn generator that memorizes the signal from the load detector 14 just before running, and generates a signal that changes depending on the load as a constant signal during running and adds it to the reference signal as shown in Fig. 3. . 1, 19 is the tacho quenerator, which outputs an output proportional to the car's speed. That is, the deviation between the output of the tacho generator 19 and the reference speed no. J turn generator 1 is negatively fed back to the torque of the hoisting motor 8.

Here, appropriate velocity and acceleration curves are shown in FIG. The appropriate speed curve 20 shown in the figure is shown with the vertical axis representing the speed of the elevator and the horizontal axis representing time t.In this case, the acceleration curve 21 has the vertical axis representing the acceleration of the elevator, and the horizontal axis representing the acceleration of the elevator. is shown as time t, similar to velocity curve 20. Also, A in the same figure.

Regions B and C will be referred to as acceleration mode, constant speed mode, and deceleration mode, respectively.

Next, a conventional control system will be explained based on the above-mentioned components.

- First, while the elevator is stopped, the load inside the elevator car is inputted from the load detector 14 and stored in RAM (Random Access Memory).

At the time of departure, the load pattern generator 18 outputs an offset voltage for load correction to the addition/subtraction point 3 based on this value. At the same time, the reference speed pattern I (number generator 1 is
A waveform close to the speed pattern 20 shown in FIG. 2 is output.

Here, since the tachogenerator 19 directly connected to the hoisting motor 8 provides negative feedback of an output proportional to the actual speed, the deviation is fed back and added to the armature voltage of the hoisting DC motor 8. (However, the excitation current is controlled to be constant.) However, there is a delay here, and in reality, the acceleration etc. do not stay the same as curve 21, but have some fluctuation. This results in poor riding comfort, loss of energy, and loss of arrival time. Similarly, in the deceleration mode, the landing position deteriorates due to fluctuations in speed and acceleration, and a margin must be provided for the deceleration distance. In addition, a landing/turn generating device 13 is definitely required.

Furthermore, these factors also complicate maintenance of the device.

[Purpose of the invention]

SUMMARY OF THE INVENTION An object of the present invention is to provide an elevator control device that eliminates fluctuations in the actual speed and acceleration pattern of an elevator, improves riding comfort, reduces arrival time and energy loss, and improves landing accuracy. It is said that

[Summary of the invention]

A feature of the present invention is that a speed and acceleration reference pattern generator 5- is provided, and this reference four-turn generator 5- generates preset speed and acceleration coefficient values as the deviation between the set value and the actual value of speed and acceleration, and the car The purpose is to compensate for the internal load.

[Embodiments of the invention]

FIG. 4 shows the configuration of a control device according to an embodiment of the present invention.
The same parts as those in the figure are shown with the same reference numerals.

23 is a reference pattern generator, which generates an ideal acceleration provided inside, an acceleration value output from a speed pattern generator,
The speed value is multiplied by a coefficient corrected by the car internal load and output. Internal turn generator and position detector 6. The output of the tacho generator 19 is switched as shown in FIG. 2 based on a value proportional to the speed and an acceleration value predicted by differentiating the value. This situation will be described in detail later. Further, 24 is a dynamic coefficient correction device that corrects the coefficient to be multiplied, and corrects the coefficient based on the ratio of the ideal acceleration and velocity to the actual acceleration and velocity. Also, an offset value due to the internal load is added to the output 6- of the reference/turn generator 23. A portion 25 indicated by a broken line in the figure indicates a microcomputer. The configuration of this microcomputer is the 10th
The configuration is as shown in the figure.

Next, as a tool to explain the reference/main turn generator 23 and the dynamic coefficient correction device 24, the equation of motion of the elevator and its determination method will be shown.

These are preparations in advance, and can be measured by connecting a pen recorder or the like to the output of the tacho generator 19 and the input voltage of the Keyne converter 4 of the apparatus shown in FIG. Therefore, first we find the equation of motion of the elevator.

Figure 5 shows the main elements of the elevator's equation of motion. Inductance La, resistance Ra, inductance Lf, and resistance Rf are the inductance and resistance of the armature circuit and field circuit □ of the DC motor for the hoisting machine, respectively. voltage vB
I Vf is the input voltage of the circuit, il is the current in the field circuit, and Maf is a constant proportional to the mutual inductance between the armature winding and the field winding. Here, constant excitation control of the DC motor is performed with tl=constant. Still, X is the elevator car position (m) from the starting point.

θ is the rotation angle (rad) of the sheave 1θ, and Jθ is the total inertia moment (kyrn) of the elevator system (motor armature, counterweight 11, car 12, car internal load) with respect to θ.
) + Dθ is the overall viscous friction constant (N
ms/rad), Mm is the weight of counterweight 11 (kg), Mc is the weight of basket 12 (2), ΔM is basket 1
The load (kg) in 2 and r are the radius (m) of the sheave 10, and these are constants on the elevator main body side. here 3
ΔM of 7 is a variable depending on the number of passengers, and other installed elevators and DC motors 8. This is a constant determined by the sieve io and their axes. However, the numbers shown here are the central elements of the elevator's equation of motion.

Next, find the equation of motion of the elevator using the values shown here. Mass M corresponding to unbalanced load,
lTotal mass MP is expressed as (]) + (2
).

MP=Mc10Mm〔Yu〕 ・・・・・・・・・・・・
・・・・・・・・・・・・・・・・・・(1) M8-M
0+MmCkg〕 ・・・・・・・・・・・・・・・
・・・・・・・・・・・・・・・・・・(2) When rearranged in this way, the equation of motion is shown by the following equation.

(J, 9+2πr(Mp+ΔM):]'+Dθ#+(M
s+ΔM)rr (where a, 'o' are the angular velocity and angular acceleration of the sheave. Also, 2 is the gravitational acceleration,
Approximately 9.8 (m/82) The left side of this equation (3) indicates the load torque external to the motor,
The right side is the output torque of the electric motor. In order to transform the equation (3) from a system based on the sheave rotation θ to a system based on X related to the position, 9-j=''...・・・・・・
- (5) Substituting 2πr yields equation (7). However, the delusions and parables are the speed and acceleration of the elevator, respectively.

To make it easier to understand, we will transform each constant as follows. However, due to constant excitation control that keeps il constant, (M
It is assumed that af'j> is a constant value.

Jθ/2πr+M P3J・・・・・・・・・・・・ (
8) 7a:llo+Δu == u・・・・・・・・・
...αη-10= (where al indicates the dinning between the dyne conversion amplifier 4 and the DC motor terminal voltage, U is the input voltage of the amplifier 4, and uQ is the input load used for balancing. (These will be formulated later.) In addition, regarding what is shown in equations (6) to (8), J is the overall moment of inertia of the X system with respect to the input U.

K can be considered as the input dyne for the input U of the X system, and D can be considered as the overall viscous friction constant for the input U of the X system.

In the formula 01), Ku 6 = (MB + ΔM
)rr・・・・・・・・・・・・・・・・・・・・・
・・・・・・・・・・・・・・・ 0■ Torrential u(
1, this can be referred to as the offset input of the unbalanced load balance. However, ΔM here is a value determined from the internal load detector. From these equations, equation (7) is simplified and expressed as the following equation 02◆.

(J + ΔM) + DM = ΔU ・,・・・・・・, Near ・・・・・・・・・・・・・・・・・・・・・ (1: For each of these constants, D, Jil− 1: Stall test during elevator installation (torque when restrained and not rotating),
Determined by constant speed test and falling test. However, these can also be calculated and verified from the original formula. The methods for these tests are shown below. The stall test is a test in a stopped state, as shown in Figure 6, and the
〇.

) (-Q, ΔM=00. Also, a tension meter is added to the bottom of the car, and the tension Fl is determined using this.

However, the in-car load detector 14 may be used as this tension meter. As a result, the motor output torque and tension F
From the balance of l, the following equation can be obtained for K.

In K-% column ・・・・・・・・・・・・・・・・・・
・・・・・・・・・・・・・・・・・・Negative ◆1 (However, ul is the motor input voltage at this time, and Fl is the tension.) Next, D is determined by a constant speed test. This calculation method is as shown in FIG.
The output of the tacho generator 19 is written to the pen recorder 44, and the speed is kept constant with ul constant. However, 63 is a gain conversion amplifier that converts the output of the tacho generator 19 into a speed value. In this way, X-0・・・・・・・・・・・・・・・・・・・・・
・・・・・・ θυ=smell (constant) ・・・・・・
・・・・・・・・・ OQΔM=0 ・・・・・・
・・・・・・・・・・・・・・・・・・・・・・・・
... It becomes a state of θ force, 0■0; (from the formula,
Using the input pressure u2 in this case, it is expressed as K(ul-uO) D-□ 0 clause 2. However, u6 is a correction input value when ΔM=0.

Next, determine J by a falling test. That situation is shown in Part 8.
As shown in the figure, the internal load ΔM is set to zero, the output of the tacho generator 19 is applied as the first input of the pen recorder 44, and the input voltage of the amplifier 40 is applied as the second input of the pen recorder. Then, the motor terminal voltage u2 is generated so that the speed becomes constant, and as shown in FIG. 9(a), t=
When to, drop this from ul to uQ. Let this changing input voltage be u3, and the speed at which it still changes! 3 to show the relationship with time t, Figures 9(a) and (
b) From this 9-13= time constant T of the fall of the change in X3 in FIG. 9(b), J is expressed by the following equation.

J=DT ・・・・・・・・・・・・・・・・・・
・・・・・・・・・・・・・・・・・・(11 This is 01
From the differential equation of Eq., the change X3 is X3 ” X2
@ J t ・・・・・・・・・・・・・・・
...This is more obvious than the fact that it becomes a kan. however,
t is t.

The elapsed time since, e is the base of natural logarithm, e#2.7182
It is 8.

J, D, and K were determined from the above results. Next, a control device that performs actual control using these coefficients will be explained. Figure 10 shows the configuration of the control device.

As inputs, P1 to P3 in FIG. 4, ie, the internal load detector output P3 (this is measured when the elevator is stopped) ΔM, and the elevator position P2 x.

The elevator speed P1 (according to the tachogenerator output) is input through dyne conversion amplifiers 62-64. Reference numeral 54 is an A/D (analog-digital) converter with a Y-G select function, and each signal of X + X + 3M is switched by the W-G select in the converter 54, and the converter 5
4, the analog value is converted into a digital value and input. A resolution of about 12 bits is sufficient for the 0 converter. RAM 52, which stores variables, ROM (read only memory) 51, which stores programs and constants, and a central CPU that performs control and calculations.
(Central control unit) 50. A bus controller 57 that controls paths based on commands and data output from the CPU 5θ is connected by data and address path lines 56 and 58 and command path lines 55. There is still a programmable interval timer 60 that counts time separately from the clock of the CPU 50, an interrupt controller 59 that controls its interrupts, and a D/D/D/R that outputs the calculated results.
A (digital to analog) converter 53 and its dyne conversion amplifier 61 are included.

Here, in order to explain pattern generation in an easy-to-understand manner, the first
Figure 1 shows the state variable diagram. 65 is an optimum acceleration/flat turn generator, 66 is a signal branch/output point, and coefficient Do6B.
, the coefficient Jo69 is a coefficient obtained by predicting the coefficient value of the elevator, and by using this and the integrator 67,
The elevator generates an output that follows the pattern of the acceleration/four-turn generator 65. The details will be shown later, but to put it simply, the characteristics of the elevator operation are included in this coefficient, and if the generator 65 generates an optimal flat turn, the elevator output will be adjusted to operate accordingly. It can occur. This can be said to be almost the same as a modified version of the elevator system in which acceleration is taken as input and input voltage U is output as output. For reference, the configuration of the elevator system is shown on the right side of FIG. 11. This is calculated using input voltage U as input and elevator position XP + elevator speed 2. The coefficient value BP is set so that the acceleration of the elevator is the internal state.
7J.

AP74. Integrator 73.76, signal addition point 70°72,
It consists of a signal extraction point 75, and the right side of this is
If ΔU is deformed to match ΔU in accordance with the output of the arm turn generator 65, the shape will be similar to that of the reference/arm turn generating section on the left side of the figure.

This operating principle will be explained in detail using the equation of motion 01 of the elevator obtained earlier.

The ij turn generator 65 will be explained.

This produces the + turn 21 shown in FIG. The interval timer 602 interrupt controller 59 shown in FIG. 10 generates an interrupt of, for example, 2 ms to the microconbeater, increases the acceleration by a fixed amount at a time, and stops adding for a fixed period of time when the acceleration reaches the maximum. , Also, after a certain period of time, it is decreased by a certain amount by subtraction. However, the maximum acceleration is stored in advance in a table in the ROM 51 in correspondence with ΔM (car internal load), and this is read and loaded according to ΔM and used. The generated acceleration pattern is integrated by a digital integrator to create an ideal speed pattern.

17- The digital integrator was created using a program using Simpson's 1/3 rule, and it was calculated as follows:
+2)) .........it becomes inertia.

However, in formula (c), X, the subscript t of X, t+1°t+2
represents the sampling time, and h represents the sampling time width. Here h = 2 m5e
c.

03 formula J: Su, Δu = (J+ΔM)/Niichifu+D/ is suitable...
・・・・・・・・・・・・・・・・・・ (A) Therefore, DO=D/K ・・・・・・・・・・・・・・・・・・
・・・・・・・・・・・・・・・e) Jo−(J+Δ
M)/K ・・・・・・・・・・・・・・・・・・・・・
(C) is calculated from Glue, J, and ΔM measured at the time of stoppage and stored in the memory in the memory. However, Do, J, Ni, etc. are corrected by the method described later and stored in the memory. Based on these, ΔU is calculated and stored on the RAM. Uo is the input for the balance and is 0.
→From the formula, u 6-(M8+ΔM) r? / K...
......(c), and ΔM on RAM 52
The data is calculated using the following and stored on the RAM 18-52. These data lengths are generally 16 bits, and when multiplied, the result is 32 bits long data, and the final output is ΔLI+u.

At the time of output, the data is sliced into 12-bit pieces. Furthermore, since a resolution of about 12 bits is sufficient for the 3M input, 12 bits are used for the A/1) and D/A converters. These outputs are input to the input amplifier of the elevator system that is the plant. However, since there are some errors and variations between the measured coefficients and the actual plant, the speed and acceleration may not be ideal.

Some errors and fluctuations occur between ν and the plant speed and acceleration Xp r Mp. Unfortunately, errors and fluctuations will occur between the ideal position and turn X and the plant position and turn Xp. In other words, it is necessary to correct the coefficient DO+JO in FIG. From the above, the plant coefficient BP, A, is calculated as A, %=■/(J+ΔM) by referring to the d1 formula.
・・・・・・・・・・・・・・・(a) B, = K/
(J+ΔM) ・・・・・・・・・・・・・・・・・・
・・・・・・・・・(Company) Also, uop is u 6 p # (Ms+ΔM)rs'/K...
・・・・・・・・・・・・・・・・・・・・・ (c).

Next, the correction device for this coefficient DO+JO will be explained. (
From formulas A) to (C), Δu = J o; + D o...
・・・・・・・・・・・・・・・・・・・・・・・・
(c) Since the speed Xp of the plant can be measured from the output of the tachogenerator, the predicted acceleration eccentricity of the ground can be obtained from this using a digital differentiator. This digital differentiator uses the three-point differential formula obtained from Taylor series expansion to
+4xp (published) - compromise (t+2) )...(g)
It is expressed as Pt 2hp. However, the subscripts indicate the time of sampling, and h indicates the time width of sampling.

Here, considering noise, it is set to h-about 20 m5ec. This is how it works.

and Xp are calculated. Here, u o = u o p ”””””crystal plane ゛°°“°
Assuming ゛゛゛01), the drive output of the gloss turn generator and the drive input of the ground are equal, ΔU−Δup ・・・・・・・・・・・・・・・・・・
・・・・・・・・・・・・・・・・・・・・・ 0. Here, considering the error in JP and DP of the plant, JP-Jo + ΔJo ・・・・・・・・・・・・・・・
・・・・・・・・・・・・ 01DP=D, +ΔDo
・・・・・・・・・・・・・・・・・・・・・・・・
... (B) From the relationship of equation (C), (Jo + ΔJO) xp + (Do + ΔDo ) M, =,
ro vine + Do x -...e3F'j. From this formula, (Jo+ΔJ, )=JoW ・・・・・・・・・・・・
・・・・・・・・・OQ! p (Do+ΔDo) -Do': ・・・・・・・・・
...... It becomes 0~p, and it can be seen that the correction of the coefficient, JO+DO, can be performed using the OQ, ζ function formula. The moxibustion obtained by this useless converter 54 is calculated by a set of methods,
Using these and the formula 04 on the RAM' 52, (Jo+ΔJo).

If (Do+ΔDo) is calculated and used in place of Jo and Do in FIG. 11, the coefficient error can be corrected. This correction is performed, for example, every 20 m5ec to ensure the stability of the elevator. However, in order to prevent oscillations in the elevator system, fluctuations in the coefficient values are sliced by a limiter of ±10% as shown in FIG. This is done programmatically. The same applies to Matari. These correct errors in the mechanical coefficients within a certain range.

The main parts of the present invention have been explained above.

Next, the control program will be explained according to the general flowcharts shown in FIGS. 13 and 14.

At the start of the program, the RAM 52 is initialized and necessary numerical values are set from the ROM 51. Next, a large number of ΔM (internal loads) are determined while the system is stopped. This is done to eliminate errors due to vibrations, etc., and the true internal load ΔM is estimated using a known digital filter. Next, each coefficient value is prepared and u'(1 is calculated. This is shown in the flowchart of FIG. 15.

Do and Jo are required according to formulas (a) to (c), but since Dob- is used in calculations for Do, it is usually set in this form. . Further, this is corrected during driving by a coefficient correction program.

Since J/h is also set in this format, the calculation of J is as follows: J o = J/h + ΔM/h ・・・・・・・・・・・・
・・・・・・・・・・・・・・・・・・・・・ It is carried out in the form of 0 peaks. Further, uo is obtained by equation (c), but in reality, only the ΔM part is multiplied, and the others are simply added. Also, in this part, the maximum acceleration -&8 and acceleration '1rate for ΔM are loaded. When the car next starts, the interval timer 60 is activated.

This timer 60 interrupts at 2m5ec. Departure initialization is performed before the first interrupt.

The purpose is to set the switching distance of the acceleration curve from the shortest planned stopping floor. This will also be reviewed on a regular basis. Figure 16 shows the position set for long run and short run.

Further, a flowchart is shown in FIG. (Long Run
It can also be determined by calculation.

X-Conflict dt About Gagoko's 01 formula Q]) The Shinzoson formula of the formula can be used twice. However, this time, we calculated it using simple calculations using numerical values preset in the ROM. When this is finished, wait for the 2 m5ec timer interrupt. This interrupt enters the vine generation routine (Figure 18)
. In order to create an appropriate acceleration pattern in a simple form as shown in FIG. 16, the increment, hold, and decrement programs are configured as shown in FIG. 18. In fact, more complexity is possible. This is because the following routine for calculating ΔU uses Simpson's 173 rule in formula (c) to calculate the value of the force from the vine. This is because it can be said that even if the generation routine of is changed, the following will not be affected. A flowchart of the routine for determining Δu, u is shown in FIG.

In order to simplify the calculation and make it simple in this part, the formula (a) is changed to Δu=(Jo +Do!!-)dt)+Do L(
M(t+1)+x(t+2))...(413) It is used after being modified as follows.Next, output of U is performed.This is performed every 2 m5ec.

2. Every 10 times the m5ec interrupt occurs, the coefficient correction routine is passed. This flowchart is shown in FIG. Read in Xp, calculate Xp from the difference of the three points, and from this, J. .

D. is corrected. Here too, as shown in the figure, the calculation is performed in a single time period. more than 2 m
The output U is generated by the 5ec interrupt. Also, the coefficient values related to the speed and acceleration of the elevator are corrected. Even when the motor reaches the stop position, the corrected coefficient value is retained and used for the next drive. This concludes the explanation of the control program.

Furthermore, the present invention can be modified and implemented in various ways without changing the gist thereof.

〔Effect of the invention〕

According to the present invention, the following effects can be obtained.

25- (1) Conventionally, due to speed tracking, acceleration was slow and ride comfort deteriorated, but in order to perform control by distributing both speed and acceleration, both ride comfort and tracking could be improved. It was considered sufficient.

(2) Since the acceleration pattern of the elevator and the associated speed, main turn, and position pattern can be determined to a certain extent and followed, the elevator can provide a high ride comfort and operate with less wasteful movement and less energy consumption. You can use the elevator.

(3) Since accurate position control can be performed, landing modes can be reduced, and loss time can be reduced. Furthermore, it is possible to simplify the structure of the landing pattern generator.

(4) To correct the mechanical coefficient of the elevator,
Long-term fluctuations can be automatically corrected, which reduces maintenance.

[Brief explanation of the drawing]

Figure 1 is a diagram showing the configuration of a conventional control device, Figure 126=2 is a diagram showing ideal speed and acceleration curves, Figure 3 is a diagram showing the output characteristics of the load pattern generator, and Figure 4 is a configuration diagram of a control device according to an embodiment of the present invention, FIG. 5 is a diagram showing each element in determining the equation of motion of an elevator, FIG. 6 is a diagram for explaining a stall test, and FIG. 7 is a diagram for explaining a stall test. is a diagram to explain the constant speed test, Figure 8 is a diagram to explain the falling test, Figures 9 (a) and (b) are diagrams showing the results of the falling test, and Figure 10 is the same. FIG. 11 is a state variable diagram of the reference/gut turn generating section and elevator system in the same embodiment; FIG. 12 is a diagram showing the characteristics of the limiter configured by a program; FIGS. 13-14 The figure is a general flowchart showing an example of the program in the control device of the same embodiment, Figures 15 and 17 to 2 () are flowcharts explaining the seven girders in the general flowchart, and Figure 16 is a flowchart for long run and short run. It is a diagram showing the relationship between the acceleration inflection line and the position display variable. 2. Relay contact for turn switching, 3. Addition/subtraction point, 4. Dine conversion amplifier (amplifier), 5.
... Zyristor Leonard control device, 6 ... position detection device, 7 ... governor, 8 ... electric motor for hoisting machine,
9... Interpol winding, 1o... sheave, 1no... counterweight, 12... cage, 13... landing pattern generator, 14... cage internal load transverse device, 15
... Voltage anti-hunt application circuit, 16... Current anti-hunt application circuit, 17... Landing signal generator, 18.
... Load pattern generator, 19... Tacho generator, 23... Reference speed, acceleration pattern generator, 2
4...Dynamic coefficient correction device, 25 river microcomputer. Applicant's agent Patent attorney Takehiko Suzue Figure 19 Figure 20 1111

Claims (1)

[Claims] The output of the car load detection device and the actual speed detection signal of the driven elevator are input to the reference/flat turn signal generator, and the preset coefficient values related to the speed and acceleration of the elevator are calculated in the reference/flat turn signal generator. The coefficient value is corrected by the output of the car load detection device, and the coefficient value is corrected by the deviation between the preset elevator speed and acceleration and the actual speed and acceleration, and a reference pattern signal is generated to obtain the expected speed and acceleration characteristics. An elevator control device characterized by having a configuration in which: