TWI785608B - Fluid machinery and control method thereof - Google Patents

Abstract

A fluid machinery is provided. The fluid machinery includes a controller, a controlled plant and an observer. The controller is configured to generate a first signal according to a difference between an actual pressure value of the fluid and a target pressure value. The controlled plant generates an output signal in response to a first difference between the first signal and a second signal. The observer includes an inverse model. The reverse model is established by performing a reverse operation on a physical model of the controlled plant. The output signal passes through the reverse model to generate a third signal. The third signal is filtered to generate the second signal.

Description

Fluid machinery and its control method

The present invention relates to a fluid machine, and in particular to a control method for the fluid machine.

Air compressor (Air compressor) refers to a machine used to compress air to increase gas pressure, which can provide power for various tools, transportation equipment, lifting equipment and snatching equipment. Therefore, air compressors are widely used in machinery manufacturing, metallurgy, shipbuilding, electronics, chemical industry, and oil and gas fields.

The air pressure in the air compressor cavity is expected to be maintained within a desired pressure band, so a pressure control mechanism is indispensable. In general, the pressure value can be controlled by eg a PI controller (proportional-integral controller) or a PID controller (proportional-integral-derivative controller). However, when the actual consumption of the client changes suddenly, the speed of the PI controller or PID controller to increase the working pressure to a higher target value is not fast enough, resulting in a continuous drop in the pressure value in the chamber .

Therefore, it is necessary to propose a solution to quickly reach the required target working pressure value when the actual usage of the client changes.

The invention provides a fluid machine and a control method thereof, which have the advantage of quickly reaching a required target working pressure value.

The fluid machine of the present invention includes a controller, a controlled plant and an observer. The controller is used for generating the first signal according to the difference between the actual pressure value of the fluid and the target pressure value. The controlled plant generates an output signal in response to a first difference between the first signal and the second signal. Observers include inverse models. The inverse model is established by performing an inverse operation on the physical model of the controlled plant. The output signal is subjected to an inverse model to generate a third signal. The third signal is filtered to generate the aforementioned second signal.

The fluid machine control method of the present invention is applicable to an air compression device. Fluid machinery includes controllers, controlled plants, and observers. The fluid machinery control method includes: the controller generates a first signal according to the difference between the actual pressure value of the fluid and the target pressure value; the controlled plant responds to the first difference between the first signal and the second signal value produces an output signal. Wherein, the output signal passes through the inverse model of the observer to generate the third signal. The inverse model is established by performing an inverse operation on the physical model of the controlled plant. Moreover, the third signal is filtered to generate the second signal.

Based on the above, the present invention estimates the amount of interference by setting an observer, and accordingly adjusts the first signal generated by the controller. Therefore, the present invention has a better traceability of the target pressure value when faced with external disturbances (such as a sudden increase in client usage), and can further improve efficiency.

FIG. 1 is a schematic block diagram of the control mechanism of the fluid machine of the present invention. In this embodiment, the fluid machine may be an air compressor. Please refer to FIG. 1 , the fluid machine 100 includes a nonlinear plant 110 , an observer 120 , a controller C, a computing unit 101 and a computing unit 103 . The nonlinear controlled plant 110 includes the calculator 102 and the controlled plant G. The disturbance observer 120 includes an inverse model G*, an arithmetic unit 104 and a filter F.

In this embodiment, the controller C may be a closed-loop controller, generally a PI controller (proportional-integral controller) or a PID controller (proportional-integral-derivative controller). The controller C is used to generate a signal S1 , that is, a flow control signal, according to the difference between the actual pressure value of the fluid (such as air) and the target pressure value. The arithmetic unit 101 is used for calculating the difference between the signal S1 and the signal S2 to generate the signal S3. The computing unit 102 is used to generate a signal S4, wherein the signal S4 is a sum of the signal S3 and an external interference component d (eg, a sudden change in client usage). The controlled plant G generates an output signal (corresponding to the actual pressure value P) in response to the signal S4. Taking FIG. 1 as an example, the motor frequency command (ie, the signal S3 ) actually presents pressure (ie, the actual pressure value P) after passing through the nonlinear controlled plant 110 . Therefore, the nonlinear controlled plant 110 may include the entire system. For example, the controlled plant G includes compressors (including compression devices, frequency converters, and motors), air cylinders, pipeline volumes, and so on. The motor frequency interference (i.e. the interference component d) may include the gas production equipment of the machine, all the gas consumption equipment of the client and other nonlinear parts. In this embodiment, the input and output of the controlled plant G can be flow rate and pressure respectively.

The signal S5 represents the detected output signal, which contains noise, represented by the arithmetic unit 103 and the noise component n. The inverse model G* is established by performing an inverse operation on the physical model of the controlled plant G. The signal S5 is passed through the inverse model G* to generate the signal S6. In this embodiment, the input and output of the inverse model G* can be pressure and flow, respectively. The arithmetic unit 104 is used to calculate the difference between the signal S6 and the signal S3 to generate the signal S7. The filter F is used for filtering the signal S7 to generate the signal S2.

FIG. 2 is a schematic diagram of the steps of the fluid machine control method of the present invention, wherein the fluid machine can be an air compressor. Please refer to FIG. 1 and FIG. 2 at the same time, the fluid machine 100 may include the aforementioned controller C, the controlled plant G and the observer 120 . Firstly, the controller C generates a first signal (equivalent to the signal S1 ) according to the difference between the actual pressure value of the fluid and the target pressure value (step S210 ). Next, the controlled factory G generates an output signal in response to the difference between the first signal and the second signal (equivalent to the signal S4, that is, the sum of the difference between the signal S2 and the signal S3 and the disturbance component d) (step S220). Finally, the controlled plant G outputs a signal (corresponding to the actual pressure value P). The sensed signal S5 (including the actual pressure value P and the noise component n mixed in the sensing process) passes through the inverse model G* of the observer 120 to generate a third signal (corresponding to the signal S6 ). Among them, the inverse model G* is established by performing an inverse calculation on the physical model of the controlled plant G, and the third signal is calculated and then filtered to generate a second signal (equivalent to signal S2) (step S230 ). The control mechanism of the fluid machine of the present invention will be described in detail below with reference to FIG. 3 .

FIG. 3 is a schematic block diagram of a control mechanism of a fluid machine according to an embodiment of the present invention. Please refer to FIG. 3 , the fluid machine 300 may include a control system 310 and a physical system 320 . In the control system 310 , the arithmetic unit 105 is used to calculate the difference between the sensed actual pressure value P (with noise component n) and the target pressure value Pc of the fluid to generate the signal Pe. The controller C is used for generating a signal Y1 (equivalent to the signal S1 in FIG. 1 ) according to the signal Pe. The arithmetic unit 101 is used to calculate the difference between the signal Y1 and the signal Y2 (equivalent to the signal S2 in FIG. 1 ). The calculation result of the arithmetic unit 101 is transmitted to the compression device E (that is, the signal Y) after passing through the limiter SA1. The purpose of setting the limiter SA1 is to prevent the calculation result of the computing unit 101 from exceeding the minimum and maximum values allowed by the compression device E, that is, to limit the computing result of the computing unit 101 to a value range. In this embodiment, the limiter SA1 may refer to a saturation limiter, and the minimum and maximum values allowed by the compression device E refer to the minimum and maximum values of the motor frequency.

The input and output of the compression device E may be motor frequency and flow rate, respectively. However, the present invention is not limited thereto. In other embodiments, the input of the compression device E may also be rotational speed, flow rate or current. The controlled plant G receives the output of the compression equipment E. Wherein, the output of the compression device E may be subject to external interference, such as changes in usage of other machines or clients. In FIG. 3, the output of the disturbed compression device E (corresponding to the signal S4 in FIG. 1) can be represented by the arithmetic unit 102 and the disturbance component d. The controlled plant G can respond to the output (flow rate, including the disturbance component d) of the compression equipment E to generate an output pressure. The output pressure can be sensed by a sensor (not shown in the figure), and the sensed actual pressure value P is transmitted to the computing unit 105 . The calculator 105 can calculate the difference between the sensed actual pressure value P and the target pressure value Pc. The calculation result of the calculator 105 will be used as the input of the controller C.

At the same time, the observer 120 receives the sensed actual pressure value P (with a noise component n). The inverse model G* in the observer 120 is established by performing an inverse operation on the physical model of the controlled plant G. The sensed actual pressure value P passes through the inverse model G* to generate the signal Yd. The arithmetic unit 104 is used for subtracting the signal Y from the signal Yd to generate the estimated interference value D1. The estimated disturbance amount D1 is an amount estimated for the disturbance component d in the input signal of the plant G to be controlled. The estimated interference D1 is filtered by the filter F to generate a signal D2. It should be noted that the sensed actual pressure value P (with noise component n) may be amplified due to the processing of the inverse model G*, so a filter F is required to suppress the noise in the estimated disturbance D1 Element. In this embodiment, the filter F may be a low-pass filter. The signal D2 is adjusted by the gain value Kd to generate a signal Y2 (equivalent to the signal S2 in FIG. 1 ).

FIG. 4 is a schematic block diagram of a control mechanism of a fluid machine according to an embodiment of the present invention. For the description of the components in FIG. 4 , reference may be made to the description of the same-named components in FIG. 3 , which will not be repeated here. Compared with FIG. 3 , the limiter SA2 is added in FIG. 4 . Please refer to FIG. 4 , the signal D3 is generated after the signal D2 is adjusted by the gain value Kd. Signal D3 generates signal Y2 (equivalent to signal S2 in Figure 1) after passing through limiter SA2. The function of the limiter SA2 is to prevent the signal Y2 from exceeding a preset value range. In this embodiment, the limiter SA2 may refer to a saturation limiter.

For example, the upper limit and the lower limit of the limiter SA2 are 60 and -60 respectively. When the input of limiter SA2 (ie signal D3) is 10, the output of limiter SA2 is 10. When the input of the limiter SA2 is 20, the output of the limiter SA2 is 20. That is to say, when the input value of the limiter SA2 is between the upper limit value and the lower limit value, the output and output of the limiter SA2 are the same. However, when the input of the limiter SA2 is 70, the output of the limiter SA2 is limited to the upper limit value, ie 60. When the input of the limiter SA2 is -70, the output of the limiter SA2 is limited to the lower limit value, namely -60. Limiter SA1 acts the same as limiter SA2.

Further, in an embodiment, the upper limit and the lower limit of the value range of the limiter SA2 are not fixed, but can be adjusted dynamically. For example, the upper limit and lower limit of the limiter SA1 are 60 and 24 respectively, and the upper limit and lower limit of the limiter SA2 are 60 and −60 respectively. The previous output of the controller C is denoted as Y1*, and the previous output of the observer 120 is denoted as Y2*. If the difference between Y1* and Y2* is higher than the upper limit of the limiter SA1 or lower than the lower limit of the limiter SA1, the limiter SA2 can automatically adjust its upper limit and lower limit. Specifically, when the difference between Y1* and Y2* is higher than the upper limit of the limiter SA1, by adjusting the upper limit and lower limit of the limiter SA2, the Y2 output by the limiter SA2 is greater than the upper limit. Y2* of secondary output (desired calculation result of Y1-Y2 decreases). For example, when Y1* is 75 and Y2* is 10, the upper limit and lower limit of the limiter SA2 can be adjusted to 60 and 10, respectively. When the difference between Y1* and Y2* is lower than the lower limit of the limiter SA1, by adjusting the upper limit and lower limit of the limiter SA2, the Y2 output by the limiter SA2 is smaller than the Y2 output last time * (expect Y1-Y2 calculation results to increase). For example, in the case that Y1* is 5 and Y2* is 10, the upper limit and the lower limit of the limiter SA2 can be adjusted to 10 and −60, respectively.

FIG. 5 is a continuation of FIG. 3 and FIG. 4 , and is a flow chart showing the steps of the filtering process performed by the filter F. As shown in FIG. Please refer to FIG. 3 , FIG. 4 and FIG. 5 at the same time, the process starts at step S401 . First, the estimated disturbance amount D1 is read in from the calculator 104 by the filter F (step S402 ). When the control mechanism is started and the filtering process is performed for the first time (step S403), the filter F directly uses the estimated interference amount D1 as the filtered value D2 after filtering (step S404) and outputs the filtered value D2 (step S405) . Afterwards, the estimated interference amount D1 is stored in the first memory D1 *, and the filter value D2 is stored in the second memory D2 *. So far the first calculation ends (step S407).

。在非第一次計算的情況下（步驟S403），濾波器F讀入其截止頻率f（與通過頻帶相關）、時間間隔的長度資訊、第一存儲器D1*的資料以及第二存儲器D2*的資料（步驟S408），並依據上述資料計算係數K（步驟S409）。具體來說，控制電路150可以計算公式(1)以得到係數K。係數K被儲存起來。後面會用到的第一比例與第二比例是基於係數K來決定的。 K=exp(

f

)              公式(1) Next, the estimated disturbance amount D1 is read in again from the calculator 104 by the filter F (step S402 ). In this embodiment, the arithmetic unit 104 periodically calculates the difference between the signal Yd and the signal Y according to a time interval. That is to say, there is a time interval between the time points at which the filter F obtains the estimated disturbance D1 twice, denoted as

. In the case of not the first calculation (step S403), the filter F reads its cut-off frequency f (related to the pass frequency band), the length information of the time interval, the data of the first memory D1* and the data of the second memory D2* data (step S408), and calculate the coefficient K according to the above data (step S409). Specifically, the control circuit 150 can calculate the formula (1) to obtain the coefficient K. The coefficient K is stored. The first ratio and the second ratio that will be used later are determined based on the coefficient K. K=exp(

f

) Formula 1)

In step S410 , the filter F performs an interpolation operation on the data in the first memory D1 * and the data in the second memory D2 * based on the coefficient K. The filter F determines the first ratio and the second ratio for performing the interpolation operation based on the coefficient K. In this embodiment, the first ratio is equal to 1 minus the coefficient K, and the second ratio is equal to the coefficient K. In other words, the sum of the first ratio and the second ratio is 1. The filter F can calculate the product of the data of the first memory D1* and the first ratio (1-K), the product of the data of the second memory D2* and the second ratio (K), and calculate the results of the aforementioned two products Add up to obtain the filtered value D2 and output it (steps S404 - S405 ). Then, also update the first memory D1* with the estimated disturbance value D1, and update the second memory D2* with the filter value D2. So far the second calculation ends (step S407).

Since the filter value D2 is calculated based on the previous estimated disturbance value and the previous filter value stored in the first memory D1* and the second memory D2*, the filter F can make the pressure control more stable (the pressure will eventually stop at a value). In detail, the existence of the observer 120 can improve the overall traceability and robustness, and the filter F is used as an auxiliary to further improve the stability.

FIG. 6 is a continuation of FIG. 4 , and is a flow chart showing the steps of processing the signal D3 with the limiter SA2 . Please refer to FIG. 4 and FIG. 6 at the same time, the process starts at step S501. The maximum output value Y2max, the minimum output value Y2min and the signal D3 are read from the limiter SA2 (step S502). Next, the limiter SA2 judges whether the signal D3 is less than or equal to the maximum output value Y2max (step S503 ). If not, take the maximum output value Y2max as the output (signal Y2) (step S506). If yes, the limiter SA2 further judges whether the signal D3 is greater than or equal to the minimum output value Y2min (step S504 ). If not, take Y2min as the output (signal Y2) (step S507). If yes, output the signal D3 (step S505 ). After the signal Y2 is output (step S508), the process ends (step S509).

FIG. 7 is a schematic diagram showing the curve of the flow rate changing with time. Please refer to FIG. 7 , the lines 601 and 603 represent changes in the traffic used by the client, that is, the target traffic (unit: CMM). Line 602 represents the air compressor discharge flow as a function of time without the use of the present invention. Line 604 represents the air compressor discharge flow as a function of time using the present invention. If the discharge flow of the air compressor is the same as the target flow, the pressure of the client will be stabilized at the target pressure value. It can be seen from FIG. 7 that the target flow rate also rises at the time point t, but in the case where the present invention is not used, the speed of the flow rate tracing back to the target flow rate is obviously slow. Relatively, in the case of using the present invention, the speed of traffic tracing back to the target traffic is obviously faster. That is to say, the target flow rate will also be increased when the customer's usage suddenly increases, but the traditional PI or PID controller cannot cope with this situation due to the slow traceability. On the other hand, the present invention has better traceability speed.

FIG. 8 is a schematic diagram showing the curve of pressure versus time. Please refer to FIG. 8 , the lines 701 and 703 represent the change of the pressure command from low frequency to high frequency. Line 702 represents the actual pressure response as a function of time without the use of the present invention. Line 704 represents the actual pressure response as a function of time using the present invention. It can be seen from Fig. 8 that in the case of low frequency, the actual pressure response can keep up with the pressure command. However, under high frequency conditions, line 702 shows that the pressure response cannot meet the variation of the high frequency pressure command. In contrast, line 704 (using the present invention) shows that the pressure response is more responsive to high frequency pressure command variations.

FIG. 9 is a schematic diagram showing the curve of pressure versus time. Referring to FIG. 9, lines 801 and 803 represent fixed pressure commands. Line 802 represents the pressure retrospective scenario for different flow variations without using the present invention. Line 804 represents the pressure trace scenario for different flow variations using the present invention. Among them, the flow variation changes from slow to fast with time. It can be seen from FIG. 9 that, compared with the case where the present invention is not used, the present invention has better pressure traceability in the face of different flow variations.

To sum up, the present invention estimates the amount of interference by setting an observer, and accordingly adjusts the first signal generated by the controller. Therefore, in the case of external disturbances (such as a sudden increase in client usage), since the present invention has better traceability for the target pressure value, the robustness of the control system is improved, and the control performance can also be improved. The present invention is not limited to pressure control applications, and can also be applied to applications such as valve opening control, rotational speed control, and current control.

100: Fluid Machinery 101~105: calculator 110: Nonlinear Controlled Plant 120: Observer 300: Fluid Machinery 310: Control system 320: Physical Systems 601~604, 701~704: line C: Controller d: interference component D1: Estimated amount of interference D2: filter value D3: signal E: compression equipment F: filter G: Controlled factory G*: reverse model Kd: gain value n: noise component P: actual pressure value Pc: target pressure value Pe: signal S1~S7: signal S210~S230, S401~S410, S501~S509: steps SA1, SA2: Limiter t: point in time Y1, Y2: signal Y, Yd: signal

FIG. 1 is a schematic block diagram of the control mechanism of the fluid machine of the present invention. FIG. 2 is a schematic diagram of the steps of the fluid machine control method of the present invention. FIG. 3 is a schematic block diagram of a control mechanism of a fluid machine according to an embodiment of the present invention. FIG. 4 is a schematic block diagram of a control mechanism of a fluid machine according to an embodiment of the present invention. FIG. 5 is a continuation of FIG. 3 and FIG. 4 , and is a flow chart showing the steps of the filtering process performed by the filter F. As shown in FIG. FIG. 6 is a continuation of FIG. 4 , and is a flow chart showing the steps of processing the signal D3 with the limiter SA2 . FIG. 7 is a schematic diagram showing the curve of the flow rate changing with time. FIG. 8 is a schematic diagram showing the curve of pressure versus time. FIG. 9 is a schematic diagram showing the curve of pressure versus time.

100: Fluid Machinery 101~104: calculator 110: Nonlinear Controlled Plant 120: Observer C: Controller d: interference component F: filter G: Controlled factory G*: reverse model n: noise component P: output signal S1~S7: signal

Claims (10)

A fluid machine, comprising: a controller (C), used to generate a first signal (S1) according to the difference between an actual pressure value (P) and a target pressure value (Pc) of a fluid; a a controlled plant (G) generating an output signal (P) in response to a first difference (S3) between the first signal (S1) and a second signal (S2); and an observer (120) (observer), comprising: an inverse model (G*), established by performing an inverse operation on the physical model of the controlled plant (G); a first computing unit (104), periodically based on a time interval Calculating the difference (S7) between the third signal (S6) and the first difference (S3) in order to obtain a previous estimate of the interference component in the input signal of the controlled plant (G) Measured interference and a current estimated interference; and a filter (F), used to: perform the filtering process according to a pass frequency band, the time interval and the previous estimated interference, and update the current estimate with the processing result Measure the interference amount to generate an updated current estimated interference amount, thereby generating the second signal (S2); wherein, the output signal (P) generates a third signal (S6) through the inverse model (G*) ), and the third signal (S6) is calculated and then filtered to generate the second signal (S2). The fluid machine as described in Claim 1, wherein the observer further includes: a limiter (SA2), used to limit the calculation result to a value interval after the current estimated disturbance is subjected to a gain calculation after the update , to generate the second signal (S2). The fluid machine as claimed in claim 1 or 2, wherein the filter (F) is further used to: perform an exponential operation on the pass frequency band and the time interval to generate a coefficient; and based on the coefficient, according to the previous estimate An interpolation operation is performed on the interference quantity and an updated previously estimated interference quantity to generate the processing result of the filtering process. The fluid machine as claimed in claim 3, wherein the filter (F) is further used to: perform the exponential operation according to the mathematical formula exp(-2πfΔT) to obtain the coefficient, wherein f represents the pass frequency band, and ΔT represents the time interval; performing the interpolation calculation to sum up the previously estimated interference amount of the first scale and the updated previously estimated interference amount of the second scale, thereby obtaining a processing result of the filtering process, wherein , the second ratio is equal to the coefficient and the sum of the first ratio and the second ratio is 1. A fluid machine, comprising: a controller (C), used for an actual pressure value (P) of a fluid and a target A first signal (S1) is generated by the difference between the pressure values (Pc); a controlled plant (G) responds to a first signal (S1) between the first signal (S1) and a second signal (S2) A difference (S3) produces an output signal (P); and an observer (120) (observer), including: an inverse model (G*), through a physical model of the controlled plant (G) Inverse operation is established, wherein, the output signal (P) generates a third signal (S6) through the inverse model (G*), and the third signal (S6) is calculated and then subjected to a filtering process to generate The second signal (S2); a first arithmetic unit (104), coupled to the filter (F), for calculating the third signal (S6) and the first difference (S3) at a first time point ) between the difference (S7) to obtain a first estimated interference amount for the interference component in the input signal of the controlled plant (G); and a filter (F) to obtain the first estimated interference amount The interference amount is used as a first filtering result, wherein the first filtering result undergoes a gain operation to generate the second signal (S2), the first estimated interference amount is stored to update data in a first memory, and the The first filtering result is stored to update data in a second memory, wherein the first computing unit (104) is also used to calculate the third signal (S6) at a second time point after the first time point The difference (S7) between the first difference (S3) to obtain a second estimated interference amount for the interference component in the input signal of the controlled plant (G), wherein the first time point There is a time interval between the second time point, Wherein, the filter (F) is also used to perform a filtering process according to a passing frequency band, the time interval, the data in the first memory and the second memory at the second time point, so as to generate a second filtering result , wherein the second filtering result undergoes the gain operation to generate the second signal (S2), the second estimated interference amount is stored to update the data in the first memory, and the second filtering result is stored to update a data in the second memory. A fluid machinery control method, suitable for an air compression device, wherein the fluid machinery includes a controller, a controlled plant and an observer, the fluid machinery control method includes: the controller according to an actual fluid The difference between the pressure value and a target pressure value generates a first signal; and the controlled plant responds to a first difference between the first signal and a second signal to generate an output signal, wherein , the output signal generates a third signal through an inverse model of the observer, the inverse model is established by performing an inverse operation on the physical model of the controlled plant, and the third signal is processed and then passed through A filtering process to generate the second signal, wherein the observer further includes a first computing unit and a filter, and the control method of the fluid machine further includes: periodically calculating the first computing unit according to a time interval The difference between the third signal and the first difference, to sequentially obtain the input signal for the controlled plant A previously estimated interference amount and a current estimated interference amount of the interference component; and the filtering process is performed by the filter according to a pass frequency band, the time interval, and the previously estimated interference amount, and the current is updated with the processing result Estimate the interference amount to generate an updated current estimated interference amount, thereby generating the second signal. The fluid machine control method as described in Claim 6, wherein the observer further includes a limiter (SA2), and the fluid machine control method further includes: performing a gain operation on the updated current estimated disturbance, And the calculation result is limited to a value interval by the limiter, so as to generate the second signal. The fluid machine control method as described in Claim 6, wherein the filtering process includes: performing an exponential operation on the passing frequency band and the time interval to generate a coefficient; based on the coefficient, according to the previously estimated interference amount and An interpolation operation is performed on an updated previously estimated interference quantity to generate a processing result of the filtering process. The control method of fluid machinery as described in Claim 8, wherein the step of filtering processing further includes: performing the exponential operation according to the mathematical formula exp(-2πfΔT) to obtain the coefficient, where f represents the pass frequency band, ΔT representing the time interval; performing the interpolation calculation to sum up the previously estimated interference quantity of a first scale and the updated previous estimated disturbance quantity of a second scale, thereby obtaining a processing result of the filtering process, Wherein, the second ratio is equal to the coefficient and the sum of the first ratio and the second ratio is 1. A fluid machine control method, wherein the fluid machine includes a controller, a controlled plant and an observer, the fluid machine control method includes: the controller according to an actual pressure value and a target pressure value of a fluid and a first signal is generated by the controlled plant in response to a first difference between the first signal and a second signal, wherein the output signal undergoes the observation An inverse model of the device generates a third signal, the inverse model is established by performing an inverse operation on the physical model of the controlled plant, and the third signal is calculated and then subjected to a filtering process to generate the first Two signals, wherein the observer further includes a first computing unit and a filter, and the control method of the fluid machine further includes: at a first time point: computing the third signal and the first computing unit by the first computing unit the difference between the differences to obtain a first estimated interference amount for the interference component in the input signal of the controlled plant; and use the first estimated interference amount as a first filter by the filter As a result, wherein the first filtering result is subjected to a gain operation to generate the second signal, the first estimated interference quantity is stored to update a first memory, and the first filtering result is stored to update a second data in memory; At a second time point after the first time point: the difference between the third signal and the first difference is calculated by the first computing unit to obtain the interference in the input signal for the controlled plant A second estimated interference quantity of a component, wherein there is a time interval between the first time point and the second time point; the filter is based on a pass frequency band, the time interval, the first memory and the second The data in the memory is subjected to a filtering process to generate a second filtering result, wherein the second filtering result is subjected to the gain operation to generate the second signal, and the second estimated interference amount is stored to update the first memory data, and the second filtering result is stored to update data in a second memory.

Images