JP2015186287A - Control device for dynamo electric motor and combustion apparatus with the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve follow-up performance for an actual number of revolutions with respect to a change in the target number of revolutions by exerting feed forward control coping with a change in a static characteristic of a dynamo electric motor according to individual difference of the dynamo electric motor or a change in the installation thereof.SOLUTION: According to a primary function model indicating a static characteristic of a fan motor 42, a feed forward control section 220 calculates an amount Uff of feed forward control corresponding to a target number N*of revolutions. The primary function model is determined as a predetermined reference amount of input for control input Uv and as a primary function that passes through a predetermined reference number of revolutions in the reference amount of input, in a coordinate indicating the number N of fan revolutions with respect to an amount Uv of control input to the fan motor 42. On the basis of the actual values of the amount Uv of control input to the fan motor 42, the detected number Nd of revolutions of the fan motor 42, the reference amount of input, and the reference number of revolutions, a learning section 230 learns in sequence the inclination of a primary function model used for feed forward control.

Description

  The present invention relates to an electric motor control device and a combustion device including the electric motor control device, and more particularly to motor speed control.

  In the rotational speed control of the fan motor of the combustion device, performing feedforward control based on the target rotational speed is disclosed in Japanese Patent Nos. 2526737 (hereinafter, Patent Document 1) and Japanese Patent No. 2560953 (hereinafter, Patent Document 2). Patent No. 3060728 (hereinafter referred to as Patent Document 3), Patent No. 3282169 (hereinafter referred to as Patent Document 4), and the like.

  According to Patent Document 1, according to control results in which feedback control at the pre-purge rotation speed is turned off in the control combining the feedforward control and the feedback control for setting the operation amount based on the relationship between the target rotation speed and a plurality of constants. It describes that the constant of the feedforward control is changed.

  Patent Documents 2 to 4 describe that Ford forward control is corrected based on the control results of feedback control.

Japanese Patent No. 2526737 Japanese Patent No. 2560953 Japanese Patent No. 3060728 Japanese Patent No. 3282169

  As described in Patent Document 1, in motor control of a combustion fan, feedforward control can be executed according to a linear function expression including a target rotational speed and a plurality of constants according to the static characteristics of the electric motor. However, the actual characteristics of this linear function equation vary depending on the installation status such as the individual difference of the motors that drive the fan and the blockage status of the path where the fan is installed.

  According to Patent Document 1, the constant in the linear function equation can be updated so as to absorb the error of the linear function equation due to the secular change based on the results of the pre-purge rotation speed. Here, the error of the linear function equation (y = ax + b) includes an offset error (b error) and a tilt error (a error).

  However, according to Patent Document 1, an update based on the performance at a specific rotation speed (pre-purge rotation speed) is performed, so that an effect can be expected to absorb the offset error, while the change of the slope (a) with time is achieved. There is concern that it will be difficult to learn. As a result, it becomes impossible to cope with a change in the static characteristics of the electric motor, and there is a concern that the control performance is deteriorated with respect to a change in the target rotation speed due to a reduction in the effect of the feedforward control.

  The present invention has been made to solve such problems, and an object of the present invention is to provide feedforward control that copes with changes in the static characteristics of an electric motor in accordance with changes in individual differences of motors and changes in installation conditions. Thus, it is an object to provide a motor control device with improved control performance with respect to changes in the target rotational speed, and a combustion device equipped with the motor control device.

  An electric motor control device according to the present invention is an electric motor control device in which the rotational speed changes in accordance with an electrical input, and includes a control unit, a rotational speed sensor for detecting the rotational speed of the motor, and a learning unit. Prepare. The control unit is configured to calculate a control input amount of the electrical input in accordance with a target value of the rotational speed according to a linear function model indicating a static characteristic of the rotational speed with respect to the electrical input. The learning unit sequentially learns the slope of the linear function model used for the calculation in the control unit based on the control input amount and the rotation speed detected by the rotation speed sensor. In particular, the learning unit obtains results between a difference between a predetermined reference input amount and a control input amount, and a difference between a predetermined reference rotation speed and a detected rotation speed indicating the rotation speed at the reference input amount. Based on the ratio, the learning value of the slope is sequentially updated.

  According to the motor control apparatus, it is possible to learn the slope of the linear function model in response to changes in the static characteristics of the motor including individual differences between the motors and the installation status of the motor. Then, the control based on the learned linear function model (for example, feedforward control) can promptly follow the actual rotational speed with respect to the change in the target rotational speed of the electric motor.

  Preferably, the learning unit is configured to reduce the learning speed that reflects the performance ratio in the learning value when the rotation speed is low while increasing the learning speed when the rotation speed is high.

  If this is done, there is a concern that the calculation error of the actual gradient of the linear function model (actual ratio) will increase in the region near the reference rotation speed that is generally set in the low rotation region. can do. Thereby, it is possible to prevent the learning value from fluctuating and improve the stability of learning.

  More preferably, the learning unit stops the sequential updating while the deviation of the detected rotational speed with respect to the target value is larger than the predetermined value or until the predetermined time elapses after the target value is changed. Maintain the learning value.

  In this way, learning can be stopped during a transient control response in which the motor rotation speed follows the change in the target rotation speed, so that the actual inclination of the control response when the target rotation speed is constant. Based on the above, it is possible to stably learn the linear function model of the electric motor.

  A combustion apparatus according to the present invention includes any one of the motor control devices described above, a blower fan, and a combustion mechanism. The blower fan is driven by an electric motor controlled by the control device. The combustion mechanism burns an air-fuel mixture of fuel and combustion air supplied from a blower fan.

  According to the combustion apparatus, the rotational speed of the blower fan for supplying combustion air to the combustion mechanism can be quickly controlled by following the change in the target rotational speed. As a result, the combustion air amount can be appropriately controlled by following the change in the required heat generation amount (combustion fuel amount) in the combustion mechanism, so that the combustion state can be maintained satisfactorily.

  According to this invention, the control device for the electric motor that has improved the control performance with respect to the change in the target rotational speed by the feedforward control that copes with the change in the static characteristic of the electric motor according to the individual difference of the electric motor and the change in the installation state, and A combustion apparatus provided with the same can be provided.

It is a schematic block diagram of the hot-water supply apparatus shown as a typical example of the combustion apparatus to which the control apparatus of the electric motor according to this Embodiment was applied. It is a functional block diagram of the rotational speed control of the fan motor to which the electric motor control according to the present embodiment is applied. It is a graph which shows an example of the static characteristic model of a fan motor. It is a graph which shows the change of the static characteristic of a fan motor according to the change of the obstruction | occlusion degree of the exhaust path in FIG. It is a conceptual graph for demonstrating generation | occurrence | production of control error in case the linear function model which shows the static characteristic of a fan motor has an error. It is a conceptual diagram for demonstrating the learning method of the primary function model by the learning part shown by FIG. It is a functional block diagram for demonstrating the control structure by the learning part shown by FIG. It is a functional block diagram explaining the control structure by the feedforward control part shown by FIG. It is a flowchart explaining the process sequence of the rotation speed control by the control apparatus of the electric motor according to this Embodiment.

  Embodiments of the present invention will be described below in detail with reference to the drawings. In the following, the same or corresponding parts in the drawings are denoted by the same reference numerals, and the description thereof will not be repeated in principle.

  FIG. 1 is a schematic configuration diagram of a hot water supply apparatus including a combustion apparatus to which a control device for an electric motor according to the present embodiment is applied.

  Referring to FIG. 1, a hot water supply apparatus 100 includes a combustion can body (hereinafter also simply referred to as “can body”) 10 in which a primary heat exchanger 11, a secondary heat exchanger 21, a combustion burner 30, and the like are stored. A blower fan 40, a water inlet pipe 50, a bypass pipe 60, a hot water outlet pipe 70, and a controller 200 are provided.

  A bypass pipe 60 is disposed between the water inlet pipe 50 and the hot water outlet pipe 70. A distribution valve 80 for controlling the diversion to the bypass pipe 60 is connected to the water inlet pipe 50. Further, a temperature sensor 110 and a flow rate sensor 150 are arranged in the water intake pipe 50. The temperature sensor 110 detects the incoming water temperature Tw.

  Tap water or the like is supplied to the water intake pipe 50. Depending on the opening degree of the distribution valve 80, a part of the water supply amount is diverted from the water inlet pipe 50 to the bypass pipe 60. That is, the ratio of the divided flow rate to the total water supply amount to the water intake pipe 50 is controlled according to the opening degree of the distribution valve 80. The flow sensor 150 is disposed on the downstream side (can body side) of the distribution valve 80. Accordingly, the flow rate Q detected by the flow rate sensor 150 indicates the flow rate that passes through the can 10 (can flow rate).

  The water in the water intake pipe 50 is first preheated by the secondary heat exchanger 21 and then mainly heated in the primary heat exchanger 11. Hot water heated to a predetermined temperature by the primary heat exchanger 11 and the secondary heat exchanger 21 is output from the can 10 to the hot water discharge pipe 70.

  The outlet pipe 70 is connected to the bypass pipe 60 at the junction 75. Accordingly, hot water having an appropriate temperature obtained by mixing the hot water output from the can 10 and the water from the bypass pipe 60 is poured from the hot water supply device 100 into a hot water tap 190 such as a kitchen or a bathroom or a bath (not shown). It is supplied to a predetermined hot water supply location such as a hot water circuit.

  The outlet pipe 70 is provided with a flow rate adjusting valve 90 and temperature sensors 120 and 130. The temperature sensor 120 is disposed on the upstream side (can body side) of the outlet pipe 70 with the bypass pipe 60 and detects the output hot water temperature (hereinafter, can body temperature) from the can body 10. The temperature sensor 130 is provided on the downstream side (the hot water side) from the junction 75 and detects the hot water temperature Th after the water from the bypass pipe 60 is mixed. The flow rate adjusting valve 90 is provided to control the hot water flow rate. For example, it is possible to prevent the hot water temperature Th from being lowered by reducing the opening degree of the flow rate adjustment valve 90 during a period in which the heating capacity immediately after the start of combustion is insufficient.

  In the hot water supply apparatus 100, the combustion apparatus can be configured by the combustion burner 30 and the blower fan 40. In the can 10, the water supplied from the inlet pipe 50 is heated by the primary heat exchanger 11 and the secondary heat exchanger 21 by the amount of heat generated by the combustion device.

  In the can 10, the fuel gas output from the combustion burner 30 is mixed with the combustion air from the blower fan 40. When the air-fuel mixture is ignited by an ignition device (not shown), the fuel gas is burned and a flame is generated. The combustion heat generated by the flame from the combustion burner 30 is given to the primary heat exchanger 11 and the secondary heat exchanger 21 in the can 10.

  The primary heat exchanger 11 heats incoming water by heat exchange by sensible heat (combustion heat) of combustion gas by the combustion burner 30. The secondary heat exchanger 21 heats the water passed by the latent heat of the combustion exhaust gas from the combustion burner 30 by heat exchange. An exhaust path 15 for exhausting the exhaust gas after heat exchange is provided on the downstream side of the can body 10 in the flow direction of the combustion gas.

  In the gas supply pipe 31 to the combustion burner 30, an original gas solenoid valve 32, a gas proportional valve 33, and capacity switching valves 35a to 35c are arranged. The original gas solenoid valve 32 has a function of turning on and off the supply of fuel gas to the combustion burner 30. The gas flow rate of the gas supply pipe 31 is controlled according to the opening degree of the gas proportional valve 33.

  The capacity switching valves 35 a to 35 c are controlled to be opened and closed in order to switch the number of burners to be supplied with fuel gas among the plurality of combustion burners 30. In the example of FIG. 1, the supply of fuel gas to the five combustion burners 30 on the left side in the drawing among the ten combustion burners 30 is turned on / off according to the opening / closing of the capacity switching valve 35a. Similarly, the supply of fuel gas to the two combustion burners 30 in the center of the figure is turned on and off in response to the opening and closing of the capacity switching valve 35b, and the supply of fuel gas to the three combustion burners 30 on the right side in the figure is switched to the capacity. It is turned on / off in response to opening / closing of the valve 35c.

  The amount of heat generated from the combustion apparatus is proportional to the amount of gas supplied from the entire combustion burner 30 determined by the combination of the number of burners and the gas flow rate. Therefore, a setting map that determines the combination of the opening / closing pattern (number of burners) of the capacity switching valves 35a to 35c and the opening degree (gas flow rate) of the gas proportional valve 33 is prepared in advance in accordance with the required heat generation amount for the combustion device. Can do.

  The amount of air blown by the blower fan 40 is controlled so that the air-fuel ratio with the amount of gas supplied from the entire combustion burner 30 becomes a predetermined value (for example, the ideal air-fuel ratio). Since the amount of air blown by the blower fan 40 is proportional to the number of fan revolutions, the number of revolutions of the blower fan 40 needs to be controlled according to the target number of revolutions set according to the change in the amount of supplied gas.

  The blower fan 40 is driven by a fan motor 42 configured by an “electric motor”. Generally, an electric motor has a characteristic that the number of rotations changes according to an electric input (voltage and / or current). In the example of FIG. 1, the fan motor 42 is configured by a DC motor. That is, the rotation speed of the fan motor 42 is controlled by a fan drive voltage supplied from a power converter (not shown) to the fan motor. The rotational speed (fan rotational speed) of the blower fan 40 is detected by the rotational speed sensor 45.

  In the configuration example of FIG. 1, the fan rotation speed corresponds to the rotation speed of the fan motor 42. Therefore, in the combustion apparatus in the hot water supply apparatus shown in FIG. 1, the fan rotational speed control of the blower fan 40 is equivalent to the rotational speed control of the fan motor 42. For example, the rotation speed sensor 45 is configured by an electromagnetic pickup type sensor attached to a rotating body of the blower fan 40 or the fan motor 42.

  Controller 200 receives an output signal (detected value) from each sensor and a user operation, and generates a control command to each device in order to control the overall operation of hot water supply apparatus 100. The user operation includes an operation on / off command for hot water supply apparatus 100 and a set hot water temperature (Tr *) command. The control command includes opening / closing and opening command of each valve, and an electric input command (fan drive voltage command) to the fan motor 42.

  When the operation command for hot water supply device 100 is turned on, controller 200 turns on the combustion operation by the combustion device in response to flow rate Q detected by flow rate sensor 150 exceeding the minimum operating flow rate (MOQ). When the combustion operation is turned on, the original gas solenoid valve 32 is opened and the supply of fuel gas to the combustion burner 30 is started.

  In the hot water supply apparatus 100, heated water (temperature Tw + ΔT) from the can body 10 and non-heated water (temperature) from the bypass pipe 60 are disposed from the flow rate adjustment valve 90 disposed on the downstream side (outlet side) from the junction 75. The hot water mixed with Tw) is output. If the diversion rate to the bypass pipe 60 according to the opening of the distribution valve 80 with respect to the total water supply amount is (1−r) (0 <r ≦ 1.0), the flow rate detected by the flow rate sensor 150. Using Q (can body flow rate), the hot water flow rate from the flow rate adjusting valve 90 is indicated by Q / r.

  The controller 200 calculates a required heat generation amount P * for the combustion device when the combustion operation is on. Specifically, the target value of the can body temperature is calculated from the set hot water temperature Tr * and the incoming water temperature Tw in consideration of the diversion rate by the distribution valve 80. If the temperature rise by the combustion device is ΔT, the can body temperature becomes Tw + ΔT. Therefore, the temperature rise amount ΔT by the combustion device can be obtained by subtracting the incoming water temperature Tw from the target value of the can body temperature.

  The controller 200 calculates the required heat generation amount P * to the combustion apparatus from the flow rate Q and the temperature increase amount ΔT (P * = Q · ΔT). Further, the controller 200 sets the amount of gas supplied from the entire combustion burner 30 according to the required generated heat amount P *. Specifically, a combination of the number of burners and the gas flow rate that realizes the supply gas amount is determined according to the setting map. The controller 200 controls the opening degree of the gas proportional valve 33 and the opening / closing of the capacity switching valves 35a to 35c so that the number of burners and the gas flow rate according to the setting map are realized.

  Further, the controller 200 sets a target value N * of the fan rotation speed (hereinafter, target rotation speed N *). As described above, the target rotational speed N * needs to be set in proportion to the required generated heat amount P * from the combustion device, that is, the amount of gas supplied from the entire combustion burner 30.

  Furthermore, the controller 200 controls the fan rotation speed to the target rotation speed N * by controlling the fan motor 42 that drives the blower fan 40.

  FIG. 2 is a functional block diagram of the rotational speed control of the fan motor to which the electric motor control according to the present embodiment is applied. Note that the function of each block in each functional block diagram including FIG. 2 may be realized by software processing by execution of a predetermined program of the controller 200, and dedicated hardware (electronic Circuit) may be built in the controller 200.

  Referring to FIG. 2, controller 200 includes a deviation calculation unit 205, a feedback control unit 210, a feedforward control unit 220, a learning unit 230, and an addition unit 250.

  The controller 200 calculates the control input amount Uv of the fan motor 42 that is the control target of the rotational speed control for each control cycle. The control input amount Uv is a voltage or current supplied to the DC motor constituting the fan motor 42.

  The deviation calculating unit 205 calculates a rotational speed deviation ΔN (ΔN = N * −Nd) between the target rotational speed N * and a detected value of the fan rotational speed N (hereinafter also referred to as a detected rotational speed Nd) by the rotational speed sensor 45. Calculate. The feedback control unit 210 calculates a feedback control amount Ufb according to the rotation speed deviation ΔN. For example, the feedback control unit 210 is configured by at least a part of P (proportional) control, I (integral) control, and D (differential) control with respect to the rotational speed deviation ΔN.

  The feedforward control unit 220 calculates a feedforward control amount Uff according to the target rotational speed N *. The adder 250 calculates the control input amount Uv of the fan motor 42 according to the sum of the feedforward control amount Uff and the feedback control amount Ufb.

  As described above, the controller 200 calculates the control input amount Uv so that the detected rotational speed Nd matches the target rotational speed N * by a combination of feedforward control and feedback control.

  Here, in the rotational speed control, transient response control for promptly following the detected rotational speed Nd to the change in the target rotational speed N * and steady control for stabilizing the detected rotational speed Nd with respect to a constant target rotational speed N *. Both are required. Since the feedback control is based on the rotational speed deviation ΔN, it mainly contributes to steady control. On the other hand, the transient response control is realized mainly by feedforward control.

  As described above, in hot water supply device 100, target rotational speed N * varies according to required generated heat amount P * in the combustion device (that is, the amount of gas supplied from the entire combustion burner 30). For this reason, the target rotation speed N * is not maintained constant during one combustion operation, and changes according to changes in the flow rate Q and the like. Therefore, the importance of the feedforward control for dealing with the change in the target rotational speed N * is increased.

  The feedforward control unit 220 calculates a feedforward control amount Uff according to a linear function model indicating the static characteristics of the fan motor 42.

  FIG. 3 is a graph showing an example of a static characteristic model of the fan motor 42. FIG. 3 shows a Uv-N coordinate plane in which the horizontal axis represents the control input amount Uv of the fan motor 42 and the vertical axis represents the fan rotational speed N.

  Referring to FIG. 3, fan rotation speed N changes according to a linear function model 500 on the Uv-N coordinate plane expressed by the following equation (1) with respect to a change in control input amount Uv to fan motor 42. To do.

N = k · Uv + b (1)
In the equation (1), the slope k of the linear function is represented by the change amount ΔN of the fan rotation speed with respect to the change amount ΔUv of the control input amount Uv to the fan motor 42. b is a constant term.

  By formulating the linear function model 500 according to the actual machine experiment result, the feedforward control amount Uff is changed to k · ΔN according to the change amount ΔN * of the target rotation speed N * according to the change in the target rotation speed N *. * Only can be changed immediately.

  Accordingly, when an error occurs in the linear function model 500, the change amount (k · ΔN *) of the feedforward control amount Uff becomes excessive or insufficient, so that the fan rotation speed N quickly changes to the target rotation speed N *. It becomes impossible to follow. For example, there is a concern that the linear function model 500 shown in FIG. 3 has an error with respect to an actual static characteristic due to the influence of individual differences of motors, changes in installation conditions, and the like.

  In the application example to the hot water supply apparatus 100 shown in FIG. 1, static characteristics (control input amount Uv−) of the fan motor 42 are changed by changing the air resistance with respect to the rotation of the blower fan 40 due to adhesion of soot in the exhaust passage 15. There is concern that the fan speed N) will change. This static characteristic changes in accordance with the degree of soot adhesion, that is, the degree of blockage of the exhaust passage 15.

  FIG. 4 shows a graph showing changes in the static characteristics of the fan motor in response to changes in the degree of blockage of the exhaust passage 15. In FIG. 4, the horizontal axis represents the control input amount Uv of the fan motor 42, and the vertical axis represents the detected rotational speed Nd.

  Referring to FIG. 4, the static characteristics of fan motor 42 are obtained by sampling detected rotation speed Nd while changing control input amount Uv while the degree of blockage of exhaust path 15 is constant, and using the least square method as measurement data. And so on. The degree of blockage of the exhaust passage 15 is set to a plurality of levels, and the static characteristics obtained at each level are shown as characteristic lines 500 to 504 in FIG.

  In FIG. 4, a characteristic line 500 corresponds to a linear function model when a new product is used, that is, when the exhaust passage 15 is not blocked. On the other hand, the characteristic lines 501 to 504 correspond to a linear function model when the exhaust passage 15 is blocked. The degree of blockage of the exhaust passage 15 increases in the order of the characteristic lines 501, 502, 503, and 504.

  As the degree of blockage increases, the amount of outside air introduced in accordance with the rotation of the blower fan 40 decreases. For this reason, the fan rotation speed with respect to the same control input amount Uv increases as the air resistance decreases. Such a change in the static characteristics due to the blockage of the exhaust passage 15 is shown as a representative example of the secular change due to the installation situation or the like. Alternatively, the actual static characteristics of the fan motor 42 may have an error with respect to the linear function model 500 from the beginning of a new product due to individual differences of the fan motor 42.

  FIG. 5 shows a conceptual graph for explaining the generation of a control error when the linear function model has an error.

  FIG. 5 shows a control operation when the feedforward control is executed according to the linear function model 500 in a state where the actual static characteristic of the fan motor 42 is indicated by the characteristic line 505 (FIG. 5). In FIG. 5, the horizontal axis represents the fan rotation speed N, and the vertical axis represents the control input amount Uv of the fan motor 42.

  Referring to FIG. 5, in the state where target rotational speed N * = N1, fan rotational speed N = N1 is controlled along control line amount Uv = U1 along actual characteristic line 505. Therefore, in the feedforward control according to the linear function model 500, the control input amount Uv is insufficient by ΔU1. This deficiency ΔU1 is compensated by the feedback control amount Ufb.

  Consider a case where the target rotational speed N * has changed to N2 from this state. In this case, feedforward control amount Uff increases according to characteristic line 500 # in which the slope of linear function model 500 is equal. Characteristic line 500 # corresponds to linear function model 500 shifted by ΔU1.

  On the other hand, in order to control the fan rotational speed N = N2, it is necessary to set the control input amount Uv = U2 in accordance with the actual characteristic line 505. For this reason, in the feedforward control alone, the control input amount Uv is insufficient by ΔU2, so that a rotational speed deviation corresponding to ΔU2 occurs. This rotational speed error is compensated by feedback control according to the rotational speed deviation.

  The feedback control mainly compensates for a steady deviation by a proportional term and an integral term. In particular, the integral term is effective in compensating for the offset error and eliminating the steady deviation, but since the phase delay inevitably exists, the control target value (here, the target rotational speed N *) In a control system that frequently changes, it is difficult to increase the control gain.

  Therefore, in a control system in which the control target value (here, the target rotational speed N *) changes frequently, such as the rotational speed control of the fan motor 42 in the hot water supply apparatus 100 shown in FIG. is important. In the motor control according to the present embodiment, the learning unit 230 shown in FIG. 2 learns a linear function model expression based on the actual operating state of the fan motor 42, thereby improving the accuracy of the feedforward control.

  Referring again to FIG. 2, the learning unit 230 learns the learning value kL (inclination k of the linear function model 500 based on the control input amount Uv of the fan motor 42 and the actual result of the fan rotational speed N (detected rotational speed Kd). Hereinafter, the slope learning value kL) is sequentially calculated.

  FIG. 6 is a conceptual diagram for explaining a learning method of the linear function model by the learning unit 230. FIG. 6 shows the Uv-N coordinate plane as in FIG.

  With reference to FIG. 6, in the motor control according to the present embodiment, a constant reference point 550 that always passes even if inclination k changes is set on linear function model 500 of fan motor 42. The reference point 550 is defined by a combination of the reference input amount U0 and the reference rotation speed N0 on the Uv-N coordinate plane.

  From the linear function model 500 and the characteristic lines 501 to 504 shown in FIG. 4, the reference points (U0, N0) at which the fan rotational speed N with respect to the same control input amount Uv can be regarded as almost the same in the low rotational speed region. It is understood that it can be determined by actual machine experiments.

  Therefore, as shown in FIG. 6, the learning unit 230 defines the linear function model 500 as a linear function passing through the reference point (U0, N0), and determines the actual control input amount Uv and the fan rotational speed N. From this, the inclination k is learned. For example, in the initial state of the linear function model 500, the learning value kL = k0 (initial value) of the slope k. In a state where the static characteristic of the fan motor 42 changes from the initial state and the fan rotation speed N changes according to the characteristic line 505, the learning unit 230 updates the learning value kL = k1. Therefore, by performing feedforward control using the updated learned value kL, the errors ΔU1 and ΔU2 of the control input amount Uv shown in FIG. 5 can be reduced.

FIG. 7 is a functional block diagram for explaining a control configuration by the learning unit 230.
Referring to FIG. 7, learning unit 230 includes an inclination calculation unit 232 and an update processing unit 235. The gradient calculation unit 232 sequentially calculates the gradient k of the linear function model passing through the reference point (U0, N0) based on the actual control input amount Uv and the detected rotation speed Nd.

  For example, in the nth (n: natural number) control cycle, the slope calculation unit 232 follows the following equation (2), and the slope k (n− of the linear function model in the previous (n−1) th control cycle. The actual value of 1) is calculated.

k (n-1) = (Nd (n-1) -N0) / (Uv (n-1) -U0) (2)
In equation (2), Nd (n−1) and Uv (n−1) are the detected rotation speed Nd and the control input amount Uv of the fan motor 42 in the (n−1) th control cycle, respectively. Show. That is, in each control cycle, the inclination calculation unit 232 calculates the difference between the reference input amount U0 (constant) and the actual control input amount Uv, and the difference between the reference rotation number N0 (constant) and the detected rotation number Nd. Are sequentially calculated as actual values of the slope of the linear function model.

  The update processing unit 235 sequentially updates the gradient learning value kL by smoothing the time series change of the gradient k sequentially obtained in each control cycle.

  For example, the update processing unit 235 updates the gradient learning value kL by reflecting the newly calculated actual gradient value k (n−1) with respect to the previous gradient learning value kL by low-pass filter processing. To do. For example, the update processing unit 235 calculates an inclination learning value kL (n) in the nth control cycle according to the following equations (3) and (4).

kL (n) = α · k (n−1) + (1−α) · kL (n−1) (3)
α = 1 / (L + 1), (1-α) = L / (L + 1) (4)
L shown in Expression (4) is a parameter for adjusting the learning speed in the update processing unit 235, and L ≧ 0. When L = 0, α = 1, so that the actually calculated gradient actual value k (n−1) is directly used as the gradient learning value kL (n) (kL (n) = k (n−1). ). That is, the learning speed is a maximum value. On the other hand, if L = ∞, α = 0, and the inclination learning value kL (n) does not change. That is, learning is stopped.

  Thus, as the adjustment parameter L increases, α decreases and the ratio of reflecting the actual tilt value to the learned learning value decreases. That is, the learning speed is lowered. On the other hand, as the adjustment parameter L becomes smaller, α becomes larger and the ratio of reflecting the actual inclination value to the learned learning value becomes higher. That is, the learning speed is increased.

  The adjustment parameter L may be a constant value. It may be changed according to the state of the fan motor 42. For example, in a low rotation region close to the reference point 550, there is a concern that the detection error of the actual tilt value becomes high. Therefore, it is preferable to reduce the learning speed by reducing the adjustment parameter L. On the contrary, in the high rotation region, it is preferable to increase the learning speed by increasing the adjustment parameter L. Thus, the adjustment parameter L can be changed according to the detected rotation speed Nd.

  In addition, it is preferable in terms of accuracy that the static characteristic (linear function model) of the fan motor 42 is learned in a state where the rotation speed is stable. That is, it is preferable that learning of the linear function model is not executed during transient response control in which the fan speed N follows the change in the target speed N *. For example, the learning unit 230 determines the linear function model until the predetermined time elapses after the target rotational speed N * changes or while the rotational speed deviation magnitude | ΔN | is larger than a predetermined value. It is preferable to stop learning. Since the learning unit 230 stops updating the gradient learning value kL for each control period when learning is stopped, the gradient learning value kL is maintained.

  FIG. 8 is a functional block diagram illustrating a control configuration by the feedforward control unit 220 shown in FIG.

  Referring to FIG. 8, feedforward control unit 220 includes a subtracting unit 222, a multiplying unit 223, and an adding unit 224. The subtracting unit 222 subtracts the reference rotational speed N0 corresponding to the reference point 550 (FIG. 6) from the target rotational speed N *. The subtraction unit 222 outputs (N * −N0) obtained by the calculation to the multiplication unit 223.

  The multiplication unit 223 multiplies the (N * −N0) from the subtraction unit 222 by the gradient learning value kL obtained by the learning unit 230. In the nth control period, the multiplication unit 223 outputs the learning value kL (n) · (N * (n) −N0) to the addition unit 224.

  The adder 224 adds kL · (N * −N0) from the multiplier 223 and the reference input amount U0 corresponding to the reference point 550 (FIG. 6) to calculate the feedforward control amount Uff.

  As described above, according to the feedforward control unit 220, the feed according to the target rotational speed N * according to the linear function model according to the reference point 550 (FIG. 6) and the learned learning value kL (n) sequentially learned by the learning unit 230. The forward control amount Uff can be set. As a result, by suppressing the control error caused by the error of the linear function model described with reference to FIG. 5, the followability of the fan speed N with respect to the change in the target speed N * can be improved. That is, the feedforward control unit 220 corresponds to an example of a “control unit”.

  FIG. 9 is a flowchart illustrating a processing procedure for rotational speed control in the motor control device according to the present embodiment. The control processing shown in FIG. 9 is executed for each control cycle as fan rotation speed control by the controller 200. FIG. 9 shows a control processing procedure in the nth control cycle.

  Referring to FIG. 9, in step S <b> 100, controller 200 acquires target rotational speed N * (n) and detected rotational speed Nd (n) in the nth control cycle. In step S110, the controller 200 uses the detected rotation speed Nd (n−1) and the control input amount Uv (n−1) acquired in the previous control cycle, and learns an inclination learning value kL (nL in the nth control cycle. n) is calculated according to equation (2). The processing in step S110 corresponds to the function of the learning unit 230 shown in FIGS.

  Further, in step S120, the controller 200 calculates a feedforward control amount Uff (n) by a feedforward control calculation using the inclination learning value kL (n) calculated in step S110. The processing in step S120 corresponds to the function of the feedforward control unit 220 shown in FIGS.

  In step S130, the controller 200 calculates a feedback control amount Ufb (n) by a feedback control calculation based on the deviation between the target rotation speed N * (n) and the detection frequency N (n) acquired in step S100. That is, the process in step S130 corresponds to the function of the feedback control unit 210 shown in FIG.

  In step S140, the controller 200 adds the feedforward control amount Uff (n) calculated in step S120 and the feedback control amount Ufb (n) calculated in step S130 to control the control input amount of the fan motor 42. Uv (n) is calculated. The processing in step S140 corresponds to the function of the adding unit 250 shown in FIG.

  Further, the controller 200 stores the control input amount Uv (n) calculated in step S140 for learning in the next control cycle. In the (n + 1) th control cycle, the inclination learning value kL is updated by the calculation in step S110 using the stored UV (n).

  As described above, according to the motor control according to the present embodiment, the linear function corresponds to the change in the static characteristics of the motor (fan motor 42) due to the secular change including the individual difference of the motor and the installation status of the motor. You can learn the slope of the model. Then, by executing the feedforward control based on the learned linear function model, it is possible to quickly follow the actual fan speed N with respect to the change in the target speed N *.

  In particular, in a drive motor for a blower fan for combustion air in a combustion apparatus, the target rotational speed N * frequently changes according to a change in the amount of fuel supplied in the combustion apparatus. For this reason, by applying the motor control according to the present embodiment, which has high follow-up performance with respect to changes in the target rotational speed, it is possible to appropriately control the amount of combustion air and stabilize the combustion state.

  Furthermore, in the integral control in the feedback control, since a phase delay inevitably exists, it is difficult to adjust the control gain in a control system in which the target rotational speed N * changes frequently. The electric motor according to the present embodiment In the control, the dependency on the integral feedback control can be reduced by increasing the effect of the feedforward control. Thereby, it is expected that the accuracy of the rotational speed control of the electric motor by the combination of the feedforward control and the feedback control can be improved.

  In addition, in this Embodiment, although the motor control which made the control object the ventilation fan of the combustion air in the combustion apparatus applied to the hot water supply apparatus was illustrated, application of this invention is not limited to such an example. Absent. That is, the motor control according to the present embodiment can be applied to the motor for driving the blower fan of the combustion device applied to other than the hot water supply device, or to the rotational speed control of the motor applied to other than the combustion device. Is possible. In particular, as described above, the effect of applying the motor control according to the present embodiment is high for a control system in which the target rotational speed of the motor changes frequently.

  The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

  10 can body, 11 primary heat exchanger, 15 exhaust path, 21 secondary heat exchanger, 30 combustion burner, 31 gas supply pipe, 32 source gas solenoid valve, 33 gas proportional valve, 35a, 35b, 35c capacity switching valve, 40 Blower fan, 42 Fan motor, 45 Rotational speed sensor, 50 Water inlet pipe, 60 Bypass pipe, 70 Hot water outlet pipe, 75 Junction point, 80 Distribution valve, 90 Flow rate adjustment valve, 100 Hot water supply device, 110, 120, 130 Temperature sensor, 150 flow rate sensor, 190 hot-water tap, 200 controller, 205 deviation calculation unit, 210 feedback control unit, 220 feedforward control unit, 222 subtraction unit, 223 multiplication unit, 224,250 addition unit, 230 learning unit, 232 inclination calculation unit, 235 Update processing unit, 500 linear function model (characteristic line), 501 to 505 Line, 550 reference point, L adjustment parameter (learning speed), N * target rotation speed, N fan rotation speed, N0 reference rotation speed, Nd detection rotation speed, P required generated heat, Q flow rate, Th tapping temperature, Tr * setting Hot water temperature, Tw incoming water temperature, U0 reference input amount, Ufb feedback control amount, Uff feedforward control amount, Uv control input amount, k slope (linear function model), kL slope learning value.

Claims (4)

A control device for an electric motor whose rotational speed changes according to an electrical input,
A control unit for calculating a control input amount of the electrical input according to a target value of the rotational speed according to a linear function model indicating a static characteristic of the rotational speed with respect to the electrical input;
A rotational speed sensor for detecting the rotational speed of the electric motor;
A learning unit for sequentially learning the slope of the linear function model used for the calculation in the control unit based on the control input amount and the rotation speed detected by the rotation speed sensor;
The learning unit includes a difference between a predetermined reference input amount and the control input amount, and a difference between a predetermined reference rotation number indicating the rotation number in the reference input amount and the detected rotation number. A control device for an electric motor configured to sequentially update the learning value of the inclination based on the actual performance ratio.   2. The learning unit according to claim 1, wherein the learning unit is configured to reduce a learning speed that reflects the performance ratio in the learning value when the rotation speed is low while increasing the learning speed when the rotation speed is high. Electric motor control device.   The learning unit stops the sequential update while a deviation of the detected rotational speed with respect to the target value is larger than a predetermined value or until a predetermined time elapses after the target value is changed. The motor control device according to claim 1, wherein the learning value is maintained. The motor control device according to any one of claims 1 to 3,
A blower fan driven by an electric motor controlled by the control device;
A combustion apparatus comprising a combustion mechanism for burning an air-fuel mixture of fuel and combustion air supplied from the blower fan.

Images