JP2008217055A - Power control method of multihop wireless system and power control system thereof - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a power control method that avoids wasteful operation of a slave station of a multihop wireless system to reduce battery consumption. <P>SOLUTION: The power control method is used in a plant instrumentation monitoring system in which a plant instrumentation process and a plant instrumentation monitoring and controlling device are connected together via a multihop wireless system. The method controls power by having a first step (S301 to S304) for collecting plant parameter measurement signals detected by the plant instrumentation process, and predicting the dynamic characteristic of the plant instrumentation process based on the measurement signals collected, a second step (S305) for determining whether or not the dynamic characteristic of the plant instrumentation process is in a stable condition based on the result of numerical value computation in the first step, and a third step (S306) for controlling the timing to start the slave station of the multihop wireless system in a way adapted to the period of the stable condition calculated in the second step. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a power control technique for a multi-hop wireless system related to reduction of power consumption of stations constituting the multi-hop wireless system.

  In recent years, attention has been focused on applying multi-hop wireless systems to the field of instrumentation monitoring. The multi-hop wireless system converts a physical quantity measured by sensors provided at dispersed measurement points into a measurement signal, and transmits the measurement signal wirelessly to a centralized monitoring device. In this multi-hop wireless system, as the name of multi-hop indicates, there is an obstacle between the measurement point and the centralized monitoring device, and one-to-one direct communication cannot be performed between the measurement point and the centralized monitoring device. In some cases, a plurality of relay stations are arranged, and a wide communication range is secured via these relay stations.

Conventionally, as an application example of a multi-hop wireless system, a CO ₂ concentration measurement signal transmitted from a CO ₂ gas sensor installed at each location of contaminated soil is acquired by a slave station of the multi-hop wireless system, and a relay station A monitoring system that sends data to a host computer connected to the master station via the network has been proposed (for example, see Patent Document 1).

According to the technique disclosed in Patent Document 1 described above, the CO ₂ concentration in the soil is continuously measured, and the progress of soil purification is monitored by the host computer from the continuous change in the measured CO ₂ concentration. Can be predicted. Usually, the slave station performs measurement and communication intermittently, not continuously. The intermittent operation is performed by activating a timer and executing measurement and communication processing at predetermined intervals or at random intervals.

The slave station is realized by, for example, a small computer having a CPU, a memory, a sensor signal input interface, a wireless communication interface, and a power supply unit. The computer is activated only when the measurement and communication processing is executed, and after the processing is executed, the computer enters a sleep mode and only the timer is operating. The timer in the sleep mode is driven and controlled by a clock circuit that drives the CPU. When a predetermined activation timer value set in advance is reached, the power supply unit is activated, and the computer shifts from the sleep mode to the operating state.
JP 2006-275940 A (paragraphs [0012] to [0022], FIG. 3)

By the way, since the above-described multi-hop wireless system does not require a wired communication cable, it has advantages such as high mobility and mobility, and in order to make better use of these advantages, the multi-hop wireless system is configured. The slave station is strongly desired to be battery-powered.
For this reason, there is a problem of ensuring maintainability by suppressing battery consumption and saving troubles such as long-term availability and battery replacement by intermittently collecting and communicating data by a timer operation.
In addition, in order to further reduce battery consumption, there is a problem that not only the power consumption of the slave station itself is suppressed, but also the number of activations is reduced.

However, according to the conventional multi-hop wireless system described above, when the slave station autonomously starts up by the timer operation and collects data, the target is in any state (for example, steady state or non-stationary state). It is not considered whether it is in a state). That is, the timer operation time of the slave station is set to a predetermined value, and the activation is monotonously repeated.
In particular, when the above-described multi-hop wireless system is applied to plant process instrumentation monitoring, if there is no significant change in the data collected by the slave station, the operation of the slave station becomes useless and unnecessary. May cause excessive battery consumption.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a power control technique for a multi-hop wireless system that can save unnecessary operation of a slave station and suppress battery consumption. .

  The power control method for a multi-hop wireless system according to the present invention is particularly effective in reducing the operation of a slave station and preventing the slave station from performing measurement and communication leading to useless power consumption. , Predict by formulating with model predictive control (MPC: Model Predictive Control), determine that it is in a stable state where no significant change is seen from the prediction result, and do not start the slave station in the stable state Therefore, it was decided to control the start timing of the slave station to adapt to the change.

  ADVANTAGE OF THE INVENTION According to this invention, the power control technique of a multihop radio | wireless system which suppresses useless operation | movement of a slave station and suppresses battery consumption can be provided.

  The outline of the present invention is that a measurement result of a plant parameter detected by the plant instrumentation process is mediated by communication between the plant instrumentation process and a plant instrumentation monitoring control device that monitors and controls the plant instrumentation process. , A power control method for a multi-hop wireless system composed of two or more stations that transmit in order to the plant instrumentation monitoring and control device, wherein in the first step, the plant instrumentation process measures plant parameters The measurement value is converted into a measurement signal, the multi-hop wireless system transmits the measurement signal to the plant instrumentation monitoring control device, and the plant instrumentation monitoring control device converts the measurement signal into the received measurement signal. A numerical calculation process for predicting the dynamic characteristics of the plant instrumentation process based on the numerical calculation process. And a boundary value of an allowable range set in advance to estimate a stable state period indicating that the numerical calculation processing result is within the allowable range. Within the range, the activation timing of the stations constituting the multi-hop wireless system is calculated, and the activation timer value is determined. In the third step, the plant instrumentation monitoring control device uses the determined activation timer value. Whether to transmit to a station constituting the multi-hop wireless system is determined based on a past activation timer value, the determination is made when the determined activation timer value and the past activation timer value are different When the start timer value is transmitted and the station receives the determined start timer value, the start timing of the station is controlled by the received start timer value.

FIG. 1 is a diagram illustrating an example of a system configuration of a plant instrumentation monitoring system 1 according to an embodiment of the present invention.
As shown in FIG. 1, the plant instrumentation monitoring system 1 includes a plant instrumentation monitoring control device 11 and a plant instrumentation process 12 including a detector 123 connected via a multi-hop wireless system 14. The plant instrumentation process 12 including the detector 121 and the operation unit 122 is connected via a plurality of controllers 13 in a wired manner.

The plant instrumentation process 12 includes plant parameters such as detectors 121 and 123 for detecting physical quantities such as flow rate, liquid level, pressure, temperature, and concentration, and control objects constituting the plant such as pumps and valves. And an operating device 122 for driving the components.
The plant instrumentation monitoring control device 11 detects the physical quantity of the plant parameter detected by the detector 123 via the multi-hop wireless system 14, and the physical quantity of the plant parameter received by the wireless data transmission section 114. Process data processing storage unit 112 that collects, processes and stores, and plant dynamics for predicting and calculating plant dynamics based on time-series data of physical quantities of plant parameters stored in the process data processing storage unit 112 Based on the prediction calculation result of the characteristic prediction calculation unit 115 and the plant dynamic characteristic prediction calculation unit 115, based on the determination result of the plant stable state determination processing unit 116 for determining the stable state of the plant and the plant stable state determination processing unit 116, Detector 123 corresponding to plant parameter indicating stable state Starting slave station of a multi-hop radio system 14 provided corresponding and a radio personal station activation management unit 117 which manages the (sleep mode).

In the plant instrumentation monitoring system 1 according to the above-described embodiment of the present invention, only the physical quantities of the plant parameters collected via the multi-hop wireless system 14 are used when predicting and calculating the plant dynamic characteristics. Not only that, but also physical quantities of plant parameters collected in the form of monitoring and control by wired data transmission via a controller 13 such as a fieldbus conventionally used are also used.
For this reason, the plant instrumentation monitoring control device 11 is connected not only to the components related to the multihop wireless system 14 described above, but also to the detector 121 and the operation device 122 of the plant instrumentation process 12, so-called loop control or sequence The wired data transmission unit 111 transmits a detection signal and an operation signal between the wired data transmission unit 111 that performs signal transmission with the controller 13 (input / output device) that performs control and the process data processing storage unit 112 described above. It also includes a monitoring control human interface (HMI) processing unit 113 used for displaying the detection signal and inputting the operation signal.

The wired data transmission unit 111 takes in a signal from the detector 121 via the controller 13, and outputs a setting signal that defines a control amount to the operation device 122 input from the monitoring control HMI processing unit 113 to the controller 13. . The supervisory control HMI processing unit 113 transmits a signal received by the wired data transmission unit 111 to the operation operator and receives an operation instruction value determined by the operation operator.
Between the wired data transmission unit 111 and the monitoring control HMI processing unit 113, there is a process data processing storage unit 112 for processing and storing a signal from the controller 13 and further processing and storing a setting signal to the controller 13. Is provided. The process data processing storage unit 112 includes a memory for storing time series data, and the memory associates the signal value of the time series data with the measurement time, and the set value and the set time of the control amount. It is configured to store.

On the other hand, the multihop wireless system 14 includes a master station 141, a relay station 142, and a slave station 143. The slave station 143 is connected to a detector 123 that detects physical parameters such as flow rate, liquid level, pressure, temperature, and concentration, which are plant parameters of the plant instrumentation process 12.
The slave station 143 takes in the signal from the detector 123, and the signal is sent to the wireless data transmission unit 114 of the plant instrumentation monitoring control device 11 via the relay station 142 and the master station 141.

The signal sent from the detector 123 to the wireless data transmission unit 114 is sent to the process data processing storage unit 112 and is stored as time series data together with the signal stored from the wired data transmission unit 111.
The flow of this process is the same as the flow of the detection signal from the plant instrumentation process and the operation signal to the plant instrumentation process by the conventional wired data transmission and wireless data transmission.

  Next, a configuration and method for determining the dynamic characteristics of the plant and performing power control of the multi-hop wireless system 14 will be described in detail.

FIG. 2 shows an example of the plant instrumentation process 12. A plant instrumentation process 12 shown in FIG. 2 shows a preparation process from raw materials to products in a typical chemical plant.
Here, the raw material stored in the buffer tank is heated with an evaporator, the first solvent A is removed, the temperature is adjusted with a heat exchanger, and finally the second solvent B is mixed with the dilution tank to adjust the concentration. And show the process of obtaining the product.

  Here, the configuration of the plant instrumentation process 12 shown in FIG. 2 will be described (see FIG. 1 as appropriate). LIC is a controller that controls the valve opening or pump speed for controlling the liquid level, PIC is a controller that controls the pump speed for pressure control, and TIC is a cooling water flow rate control for temperature control. The controller for operating the opening of the valve, DIC, is a controller for operating the opening of the flow rate adjustment valve of the heat medium or the opening of the flow rate adjustment valve of the second solvent B for concentration control.

That is, LIC, PIC, TIC, and DIC correspond to, for example, the controller 13 of the plant instrumentation process 12 shown in FIG. 1, and a liquid level gauge, a pressure gauge, a thermometer, a flow meter, and a concentration meter (not shown) Corresponds to the detector 121.
The controller connected to the DIC and controlling the opening degree of the flow rate adjustment valve for the solvent B corresponds to the controller 13 shown in FIG. 1, and the driver for the flow rate adjustment valve corresponds to the operating device 122. The controller FIC for controlling the flow rate of the second solvent B corresponds to the controller 13 shown in FIG. 1, and a detector 121 is connected as a flow meter. In FIG. 1, a plurality of controllers 13 exist independently, but signals from two detectors 121 (a concentration meter and a flow meter) may be multiplexed between the controllers 13. On the other hand, a plurality of concentration meters that are temporarily or additionally installed correspond to the detector 123 connected to the slave station 143 of the multi-hop wireless system 14. In this way, the concentration of the component substance in the dilution tank and the flow rate of the second solvent B are collected as measured values in the plant instrumentation monitoring control device 11 via the controller 13, and each set value is defined in them. Then, the opening degree of the flow rate adjustment valve of the second solvent B is transmitted to the plant instrumentation process 12 as an operation amount.

FIG. 3 is a flowchart showing the operation of the plant instrumentation monitoring system 1 according to the embodiment of the present invention.
Hereinafter, taking the concentration measurement of components in the dilution tank in the plant instrumentation process 12 of FIG. 2 as an example, monitoring of the detector 121 and the operating device 122 by wired data transmission and the detector 123 by wireless data transmission of the multi-hop wireless system 14 The power control of the slave station 143 of the multi-hop wireless system 14 by the predictive calculation control of the plant dynamic characteristics when monitoring the above is used together will be described in detail with reference to the flowchart shown in FIG. 3 (see FIG. 1 as appropriate).

  In the dilution tank, the component substance extracted from the raw material by the evaporator and the temperature adjusted by the heat exchanger flows, but the concentration distribution of the component substance is generated in the tank. A stirrer is used to achieve a uniform concentration distribution in the tank by mixing with the second solvent B. Here, assuming a substance component whose concentration distribution cannot be grasped by a single densitometer alone, a plurality of densitometers (detectors 121 and 123) are temporarily installed or added to perform concentration monitoring. Will be described.

First, the setting values of the concentration of the component substance in the dilution tank and the flow rate of the second solvent B are set by the operation operator by the monitoring control HMI processing unit 113 of the plant instrumentation monitoring control device 11, and the process data processing storage unit 112 is set. Acquires the set value (step S301).
These set values are given to the controller 13 from the process data processing storage unit 112 via the wired data transmission unit 111, and an opening degree command signal to the flow rate adjustment valve (operator 122) in the controller 13 is determined. Used for control calculations. The opening degree is stored as time series data in the process data processing storage unit 112, and the display is displayed on the screen by the monitoring control HMI processing unit 113 in a form such as a trend graph display, for example, to the operation operator. Be notified.

  The concentration of the component substance in the dilution tank and the flow rate of the second solvent B are detected and collected by the detector 121 and transmitted as time-series data via the wired data transmission unit 111, and the process data processing storage unit 112 is accumulated as time-series data (step S302, step S303). Further, the output is displayed on the process instrumentation monitoring control device 11 via the monitoring control HMI processing unit 113 in the form of, for example, a trend graph display, and is notified to the operation operator.

  On the other hand, the concentration signal of the dilution tank measured by a plurality of temporary or additional concentration meters (detectors 123) connected to the slave station 143 of the multi-hop wireless system 14 is the slave station 143, the relay station 142, the master station. 141 is sent to the wireless data transmission unit 114 and accumulated as time-series data in the process data processing storage unit 112 (steps S302 and S303). The display is monitored in the form of a trend graph display, for example. The control HMI processing unit 113 notifies the operation operator.

The time-series data stored in the process data processing storage unit 112 as described above is read by the plant dynamic characteristic prediction calculation unit 115, and the plant dynamic characteristic prediction calculation unit 115 performs the plant instrumentation process 12 by model prediction control (MPC). A numerical calculation process for predicting the dynamic characteristics of is performed (step S304).
Note that in model predictive control (MPC), if there is a dynamic model of the process for the basic control of determining the manipulated variable so that the controlled variable matches the set value, the future controlled variable will be based on that model. A process dynamic characteristic prediction formula is used based on the idea that a change in control amount can be predicted. Specifically, the plant dynamic characteristic prediction calculation unit 115 reads out the time series data of the signal measured by the detector 121 and the time series data of the signal operated by the operation unit 122 from the process data processing storage unit 112, Performs numerical computation to simulate the dynamic characteristics of plant processes.

  Usually, since it is extremely difficult to construct a precise physical model, an impulse response model or a step response model that is generally used is used. These directly define changes in the process amount in the time domain, and can be verified with actual plant operation data. In addition, in order to handle an integral process and an unstable process, there are some that use a transfer function model or an ARX (Auto-Regressive eXogneous Input) model. In the embodiment of the present invention described below, a plant response characteristic prediction method will be described using a step response model as an example.

Here, for example, consider a response to a stepped input of size 1 as shown in FIG. The value of each step response at each sampling time is represented by {a _i }. Even if it is assumed that this step response coefficient {a _i } converges to a constant value in a finite time, there is no practical problem. The step response coefficient after step s is a constant value, that is,
_{a i = a s for i ≧} s ··· (1)
And FIG. 4B shows an example of a measurement waveform by step response. If there is a dead time of the Td step,
a _i = 0 for i = 1, 2,..., Td (2)
But does not apply here.

Assuming that the plant dynamics are linear, and adding a stepped input of magnitude Δu (t) to the plant process at time t, the magnitude of the output at time t + j for this input is:
y _M (t + j) = a _j Δu (t) (3)
Given in. Based on the principle that a general solution of the output value at time t + j can be obtained by superimposing the outputs of the step-like inputs at each time point that goes back from the time t + j, a step response model represented by the following equation (4) is obtained. .

  Here, the section in which the output is predicted is from the current time to the P-1 step after the previous step L. Further, the target section (match section) for matching the predicted value with the reference trajectory (set value) is from step L to step P-1, and the target section (control section) for determining input is from the current time to M step. It is defined as At this time, since the control section ends before the coincidence section starts, M <L.

  When the above-described arithmetic expression (4) is expressed separately from the past input and the future input as viewed from the current time, it is expressed by the following arithmetic expression (5).

Furthermore, the input after the current time t + M is equal to u (t + M−1), ie
Δu (t + 1) = 0 for i ≧ M (6)
Assuming that, the following equation (7) is established.

  Here, the addition part of ∞ is reviewed as a realistic model. From the above equation (4), the estimated value at the current time t is

From this, when the estimated value y _M (t + j) of the calculation formula (7) is expressed as a change from the estimated value y _M (t) at the current time, the following calculation formula (9) is obtained.

  Here, for j, the matrix expression corresponding to the estimated value from the start step (L) to the end step (L + P−1) of the matching section is defined by the following arithmetic expression (10).

  here,

Furthermore, a term for correcting an error of the estimated value y _M (t + j) caused by a disturbance to the process or a modeling error is introduced. Usually, the difference between the actually measured value and the estimated value at the current time t is used as a correction term. In this case, the prediction formula is
y _P (t + j) = y _M (t + j) + y (t) −y _M (t) (11)
Given in.
The meaning of the correction term [y (t) −y _M (t)] is “the difference between the process and the model at the current time y (t) −y _M (t), that is, the influence of modeling error or disturbance. Will be kept constant in the future ”.

With respect to the prediction calculation formula shown by the above calculation formula (11), the matrix expression corresponding to the end step (L + P-1) from the start step (L) of the coincidence section is defined by the following calculation formula.
Y _P = Y _M + Y−Y _M0 (12)

である。 Where _YP is

It is.

When the manipulated variable is determined so that the set value (target value) r and the predicted calculation value y _P coincide with each other, if y _P (t + j) = r (t + j), the predicted calculation value y _P (t + j) is obtained. When there is a modeling error, it is conceivable that the manipulated variable changes more than necessary, resulting in poor control performance. In order to avoid such a situation, a method is adopted in which a target value that gradually approaches the set value is created, and the operation amount is determined so that the predicted calculation value matches the target value. The trajectory drawn by the target value is called a reference trajectory. This reference trajectory has two contradictory aspects required for the control system: control performance (responsiveness to following target values) and robustness (modeling error that the control system can tolerate or the degree of change in process characteristics) It introduces an adjustment parameter to find a compromise between the two.

  In addition to the reference trajectory described above, there is a penalty for the operation amount (which corresponds to weighting for the constraint condition in the optimization method) as a method for introducing the adjustment parameter, but the description is omitted.

  Here, the reference trajectory is defined by the following arithmetic expression (13) according to the definition of the coincidence section.

When the reference trajectory corresponding to the coincidence section is expressed in matrix, it is defined by the following arithmetic expression (14).
Y _R = αY + (I−α) r (14)

  here,

,
To match the reference trajectory Y _R in the prediction calculation value Y _P, to determine the current and subsequent operation amount Delta] u _f, determining the control law. When the square of the normal Euclidean norm is used as an index representing the closeness between the reference trajectory and the predicted value, the operation amount u (t + j) for ij = 0, 1,..., M−1 that minimizes the following evaluation function J Solving the problem of finding
J = ‖Y _P −Y _R・ ・ ・ (15)

Prediction calculation value _{Y P,} said the arithmetic expression (10) and calculating equation (12),
Y _p = Y + A _f Δu _f + A ₀ Δu ₀ (16)
It is. The evaluation function J is
J = (Y + A _f Δu _f + A ₀ Δu−Y _R ) ^T (Y + A _f Δu _f + A ₀ Δu−Y _R )
(17)

Arithmetic expression (17) is, for the minimum for Delta] u _f is a partial differential value must be a 0 condition by Delta] u _f to for J. That is,

Thus, assuming that A _f ^T A _f is regular, the operation amount Δu _f after the current time is represented by the following arithmetic expression (19).

An operation amount Delta] u _f arithmetic expression (19), are substituted into the arithmetic expression (10), the model predicted value _{Y M} can be calculated by the following arithmetic expression (20).

In the embodiment of the present invention described above, the arithmetic expression (20) for predicting the process dynamic characteristics is defined as described above.
The plant dynamic characteristic prediction calculation unit 115 predicts the concentration signal from the detector 123 by calculating the calculation formula (20). In this case, Y _M0 is the concentration value obtained from the detector 123 at the current time, and the constant matrices A _f and A ₀ are obtained by actually measuring or simulating the step response of the inflow adjusting valve opening relative to the concentration value. The obtained value, and the input value Δu ₀ up to the current time is the past manipulated variable of the inflow regulating valve opening.

All of these numerical values are obtained as a result of calculation by inputting what is stored in the process data processing storage unit 112 to the plant dynamic characteristic prediction calculation unit 115. Furthermore, in the formula (20), a portion corresponding to the present time since the operation amount Delta] u _f, when calculating the reference trajectory Y _R given by the arithmetic expression (14), it is necessary to configure the tuning parameters α It is necessary to compare the actual measurement result with the prediction result and set an appropriate value.
Based on the above description, the plant dynamic characteristic prediction calculation unit 115 can predict the concentration value detected by the detector 123 at a time earlier than the current time.
In the above description, a step response model is exemplified to show a method for predicting plant dynamics.

  Next, a method for predicting the dynamic characteristics of the plant when the impulse response model is used will be described. FIG. 5 is a diagram illustrating an example of an impulse response model of plant dynamic characteristics.

In this case, for example, a pulse-like input as shown in FIG. 5A is assumed to be added for one step with a magnitude of 1, and the value at each sampling time of the impulse response is represented by {a _i }. Even if this step response coefficient {a _i } is regarded as zero in a certain step, there is no problem in practice. Therefore, the impulse response coefficient after the s + 1 step is zero, that is,
a _i = 0 for i> s (21)
And FIG. 5B shows an example of a measurement waveform by step response. If there is a dead time of the Td step,
a _i = 0 for i = 1, 2,..., Td (22)
But does not apply here.

Assuming that the plant dynamics are linear and applying a pulsed input of magnitude u (t) to the plant process at time t, the magnitude of the output at time t + j for this input is:
y _M (t + j) = a _j u (t) (23)
Given in. Based on the principle that a general solution of the output value at time t + j is obtained by superimposing the outputs of the pulse-like inputs at each time point that goes back from the time t + j to the past, an impulse response model represented by the following arithmetic expression (24) is obtained. .

Hereinafter, according to the algorithm similar to the step response described above from the impulse response model given by the equation (24), the model predicted value from the start step (L) to the end step (L + P-1) of the coincidence section is calculated as follows: It can be expressed as equation (25).
Y _M = Y _M0 + A _f Δu _f + A _f Δu ₀ (25)

The plant dynamic characteristic prediction calculation unit 115 estimates the concentration value predicted to be detected by the detector 123 by calculating the calculation formula (25). In this case, Y _M0 is the concentration value obtained from the detector 123 at the current time, and the constant matrices A _f and A ₀ are obtained by actually measuring or simulating the step response of the inflow adjusting valve opening relative to the concentration value. The obtained value, and the input value Δu ₀ up to the current time is the past manipulated variable of the inflow regulating valve opening.
All of these numerical values are obtained from the results obtained by inputting the time series data stored in the process data processing storage unit 112 to the plant dynamic characteristic prediction calculation unit 115 and performing calculations.
Based on the above description, even if the impulse response model is used, the plant dynamic characteristic prediction calculation unit 115 can estimate the concentration value that is predicted to be detected by the detector 123 at a time earlier than the current time. it can.

  Returning to FIG. 3 (see FIG. 1 as appropriate), the concentration value of the detector 123 predicted by the plant dynamic characteristic prediction calculation unit 115 in this way is taken in by the plant stable state determination processing unit 116, and the plant stable state The determination processing unit 116 determines whether an important change is observed in the predicted density value. Here, an allowable upper limit value and an allowable lower limit value for the concentration are defined in advance, and no significant change is observed in the predicted concentration value by comparing these allowable range values with the predicted concentration values. It is assumed that the stable state is determined (step S305). When it is determined that the plant is in a stable state, the wireless slave station activation management unit 117 controls the activation timing of the slave station 143 of the multihop wireless system 14 according to the time length of the stable state ( Step S306).

  Hereinafter, the determination regarding the stable state of the plant by the plant stable state determination processing unit 116 and the control of the activation timing (activation timer value) of the slave station 143 of the multi-hop wireless system 14 by the wireless slave station activation management unit 117 will be described with reference to FIG. This will be described in detail with reference to FIG.

6 and 7 are diagrams illustrating an example of a time-series change in the predicted value of the plant dynamic characteristic. Here, a general time relationship between the time change of the detected concentration value and the activation timing of the slave station 143 of the multi-hop wireless system 14 will be described.
In FIG. 6A, the slave station 143 is activated at a constant interval of T1, for example, according to the activation timer value stored in advance in an internal memory (not shown) of the slave station 143. During startup, processing up to communication such as power management and path control is executed in the pre-processing section 611, and data collection and processing from the detector 123 is performed in the communication processing section 612, and communication to the relay station 142 is performed. Data transmission is performed, the processing is stopped, and a standby (sleep) state is entered. This series of processing is repeated at every timer interval T1.

In FIG. 6B, a curve 602 shows a time-series change in the density value (D). As described above, the slave station 143 activated at the timer interval of T1 detects density values at times t ₀ , t ₁ , t ₂ ,..., And D (t ₀ ), D (t ₁ ), D ( t ₂ ),.
Due to the operation of the slave station 143 at the timer interval of T1, the concentration value (D) that was within the allowable range (allowable upper limit value (Dmax) to allowable lower limit value (Dmin)) in the predetermined time region 621 is changed to the time region 622. In the case where the allowable upper limit (Dmax) is exceeded. That is, a time t _{c at} which D (t _c ) = Dmax exists between t ₁ and t ₂ . At the start time t ₂ of the slave station 143 immediately after the time t _c , the slave station 143 detects the density value D (t ₂ ) and detects the presence of the time t _c at which D (t _c ) = Dmax. Can do.

It should be noted that, when describing the concentration value D (t _n ), it accurately represents “the concentration value (D) measured by the slave station 143 activated at time t _n ”, and therefore it is not necessarily the time t _n on the time axis. Although it does not represent a measured value, D (t _n ) is added to a place where a black circle is attached to the curve of the density value (D) in the figure in order to simplify the explanation. The idea of rounding the curve in this way is the same in FIG.

Next, FIG. 7 shows a determination process in the plant stable state determination processing unit 116 based on the concentration value predicted by the plant dynamic characteristic prediction calculation unit 115, that is, step S305 in FIG. The specific processing contents of (determination of whether or not there is) is shown.
In this embodiment, it is defined that the predicted result is in a stable state as “a state in which the monitoring data value of the plant remains within the allowable range defined by the upper and lower limit values”. The type of will be described. However, the stable state of the plant includes, in addition to “a state that remains within the allowable range”, “a state that remains within the allowable range but vibrates at a constant or fluctuating period” and “within the allowable range”. It is possible to define various things such as “Steep fluctuations occur constantly or irregularly”. Any of them can be considered as an example included in the range of a so-called “unstable state” in which the plant is operated in a controllable state and deviates from the purpose of generating economical control efficiency.
In FIG. 7, the plant dynamic characteristic prediction calculation unit 115 obtains the concentration value after time t ₁ based on the prediction calculation after acquiring the concentration value D (t ₀ ) at time t ₀ at the current time t ₀ . A case is shown in which it is predicted that the curve changes as indicated by a curved line 702 indicated by a two-dot chain line imaginary line. Here, the actually acquired density value D (t ₀ ) is indicated by a black circle, and the predicted density values D (t ₁ ), D (t ₂ ), D (t ₃ ), D (t ₄ ), D (T ₅ ) is indicated by a white circle. The time when the prediction is performed is time t ₀ , that is, the origin of the horizontal axis of the graph.

According to this prediction, the concentration falls within the allowable range during the time region 721 prior to the current time t ₀ , but exceeds the allowable upper limit value (Dmax) in the time region 722 (outside the allowable range). It becomes a situation. That is, a time t _c where D (t _c ) = Dmax exists between time t ₄ and time t ₅ . If the slave station 143 activated from the time t ₁ to the time t ₄ , that is, at the times t ₁ , t ₂ , and t ₃ , the density values D (t ₁ ), D (t ₂ ), When D (t ₃ ) is detected, these values do not exceed the allowable upper limit value Dmax, and it can be predicted that the plant state will be in a stable state. Based on this prediction result, the plant stable state determination processing unit 116, the time between times t ₄ and time t ₁ is the integral multiple and the relationship (3 times in this case) the activation timer value T1, the 3 × T1 It is determined that a new start timer value T2 can be set.

  On the other hand, when it is predicted that the density value (D) changes as shown by the curve 703 indicated by the broken line, in order to make the time region until it exceeds the allowable upper limit value (Dmax) as a detectable region, It is determined that the activation timer value needs to remain T1. Since the validity that the start timer value remains T1 has been described with reference to FIG. 6, the description here is omitted in order to avoid duplication.

Next, control of the slave station activation timing by the wireless station activation management unit 117 (step S306) will be described. When the plant stable state determination unit 116 determines that the plant state is in a stable state and the start timer value needs to be changed, the radio station start management unit 117 performs the next timer start (at this time, the previous time The slave station 143 is started via the wireless data transmission unit 114, the master station 141, and further the relay station 142 during the preprocessing section 711 of the slave station 143. A new start timer value T2 is transmitted as setting data. Then, the slave station 143 that has received this rewrites the activation timer value from T1 to T2.
Furthermore, when the activation timer value T2 has elapsed and the slave station 143 has been activated, the plant dynamic characteristic prediction calculation unit 115 of the plant instrumentation monitoring control device 11 performs a prediction calculation, and the activation timer value T2 is reviewed and necessary. Update.

FIG. 8 shows specific processing for determining the stable state (step S305) and controlling the start timing of the slave station (step S306).
The processing shown in FIG. 8 will be described using the two-dot chain line curve 702 shown in FIG. 7 as an example.
First, it is searched how many times ahead of the cycle T1 the density value (D) is within the allowable range. Start time is _{t 0.} That is, a multiple n of the period T1 is set to 1 (step S801). _Then, by the calculation of _{t n = t 0 + T1 ×} n, obtains the n periods later time (step S802). Next, the density value D (t _n ) at that time is obtained, and it is compared and determined whether or not the density value D (t _n ) is within an allowable range (S803). If the density value D (t _n ) is within the allowable range (Yes in step S803), n = n + 1 (add 1 to n) is calculated (step S804), and the process returns to step S802 again.
On the other hand, if the density value D (t _n ) is not within the allowable range (No in step S803), the activation timer value T2 is calculated (step S805). In the case of FIG. 7, when the time is t _5, since the unacceptable, and has a n = 5. The reason why the operation expression step S805, the has a (n-2) is to determine the time (three cycles) of 7, from time t ₁ to time t _4.
Then, it is determined whether or not T2 is different from the previous activation timer value (step S806). If it is different (Yes in step S806), T2 is transmitted (step S807), and the process ends. On the other hand, if T2 is not different from the previous activation timer value (No in step S806), the activation timer value is not transmitted, and the process ends.
Here, an example of activation timing control (step S306) in the slave station will be specifically described below.
The start timer value of the slave station is normally set to the cycle T1. Then, at time _{t 0,} the timing when the activation timer value is computed as the change in T2 different from T1 (step S305), the time _{t 1} (after one cycle T1 of the time _{t 0),} starting a timer The value T2 is transmitted to the slave station. Then, the slave station that has received it, start from the time t ₁ after the time T2. That is, when the activation timer value is other than T1, the activation timer value T2 is transmitted. After that, when the activation timer value is not transmitted, the activation timing of the slave station returns to the cycle T1 again.
As another method, it is assumed that the slave station stores the activation timer value T2 received last time. If the activation timer value T2 is not changed, no transmission is performed. In that case, start-up timing of the slave station is, from the time _{t 0,} to repeat the T1, T2, T1, T2 ··· . On the other hand, when changing the activation timer value T2 to T1, T1 is transmitted. In other words, the activation timer value is transmitted only when the activation timer value T2 transmitted last time is changed.

  The operation amount given by the equation (20) is also applied to the control law calculation for determining the operation amount of the controller 13, which is executed in the process data processing storage unit 112 of the plant instrumentation monitoring control device 11 shown in FIG. Is applied, it is possible to realize control close to the controlled quantity predicted by Expression (10), and the accuracy of determining the stable state of the plant process is improved.

  As described above, according to the plant instrumentation monitoring system 1 according to the embodiment of the present invention, when the plant instrumentation monitoring control device 11 determines that the plant is in a stable state, the multi-hop wireless system 14. Since the start timing of the slave station 143 is determined in accordance with the period of the stable state based on the prediction of the plant dynamics, the multi-hop is performed only in a situation that is significant (predicted to be outside the allowable range) for the determination of the plant state. The slave station 143 of the wireless system 14 can be activated.

Further, according to the plant instrumentation monitoring system 1 according to the embodiment of the present invention, it is process automation that handles continuous physical quantities, and only the prediction model of plant dynamics is illustrated, but the flow of discrete objects is handled. Similarly, in the field of factory automation (FA), it is possible to apply power control of a multi-hop wireless system.
For example, there are cases where materials, processed products, and products are tracked in factories and plants, and a slave station is operated as a mobile unit for managing their positions. In this example, “position information” is a target of prediction, and “position does not change for a while”, “position moves soon”, “move far”, “move closer”, and the like are determined. .

  In this case, the process data processing storage unit 112 and the monitoring control HMI processing unit 113 included in the plant instrumentation control monitoring device 11 refer to the plant operation plan data designated by the operation operator, so that the slave station 143, for example, The moving position can be predicted based on the operation plan of the moving body mounted integrally with the detector 123 for position detection. For example, when the operation status of the slave station 143 is determined, the start timer value of the slave station 143 for position detection is set to be long for the slave station 143 that can be predicted as “not moving for a while” to reduce power consumption. Such power control can be performed.

  The power control method for a multi-hop wireless system according to the present invention, when applied to plant instrumentation supervisory control, activates a station so as to detect only a significant process change, on a sensor network for power saving purposes. It can also be applied as a sensor node, a mobile terminal, or the like.

It is a figure which shows the structural example of the plant instrumentation monitoring system which concerns on embodiment of this invention. It is a figure which shows an example of the plant instrumentation process shown in FIG. It is a flowchart which shows operation | movement of the plant instrumentation monitoring system which concerns on embodiment of this invention. It is a figure which shows an example of the step response model of a plant dynamic characteristic. It is a figure which shows an example of the impulse response model of a plant dynamic characteristic. It is a figure which shows an example of the time-sequential change of the predicted value of a plant dynamic characteristic. It is a figure which shows an example of the starting timing of the slave station of a multihop radio | wireless system. It is a figure which shows the specific process of determination of the stable state of a sub-station | mobile_unit shown in FIG. 7, and control of starting timing.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Plant instrumentation monitoring system 11 Plant instrumentation monitoring control apparatus 12 Plant instrumentation process 13 Controller 14 Multihop radio | wireless system 111 Wired data transmission part 112 Process data processing storage part 113 Monitoring control HMI processing part 114 Wireless data transmission part 115 Plant movement Characteristic prediction calculation unit 116 Plant stable state determination processing unit 117 Wireless slave station activation management unit 121, 123 Detector 122 Controller 141 Master station 142 Relay station 143 Slave station

Claims (12)

The communication between the plant instrumentation process that measures plant parameters and the plant instrumentation monitoring controller that monitors and controls the plant instrumentation process is mediated, and the measurement results of the plant parameters detected by the plant instrumentation process are displayed. , A power control method for a multi-hop wireless system composed of two or more stations that transmit in order to the plant instrumentation monitoring and control device,
In the first step,
The multi-hop wireless system transmits a measurement signal converted by the plant instrumentation process by measuring a plant parameter to the plant instrumentation monitoring and control device,
The plant instrumentation monitoring and control device performs numerical calculation processing for predicting dynamic characteristics of the plant instrumentation process based on the received measurement signal,
In the second step,
The plant instrumentation monitoring and control device calculates a start timing of a station constituting the multi-hop wireless system based on the numerical calculation processing result, determines a start timer value,
In the third step,
The plant instrumentation monitoring control device determines whether or not to transmit the determined activation timer value to a station constituting the multi-hop wireless system based on a past activation timer value, and the determined activation timer Send value,
If the station receives the determined activation timer value, the activation timing of the station is controlled by the received activation timer value;
A power control method for a multi-hop wireless system. The communication between the plant instrumentation process that measures plant parameters and the plant instrumentation monitoring controller that monitors and controls the plant instrumentation process is mediated, and the measurement results of the plant parameters detected by the plant instrumentation process are displayed. , A power control method for a multi-hop wireless system composed of two or more stations that transmit in order to the plant instrumentation monitoring and control device,
In the first step,
The multi-hop wireless system transmits a measurement signal converted by the plant instrumentation process by measuring a plant parameter to the plant instrumentation monitoring and control device,
The plant instrumentation monitoring and control device performs numerical calculation processing for predicting dynamic characteristics of the plant instrumentation process based on the received measurement signal,
In the second step,
The plant instrumentation monitoring and control device compares the numerical calculation processing result with a boundary value of a preset allowable range, and indicates a stable state period indicating that the numerical calculation processing result is within the allowable range And within the range of the estimated steady state period, calculate the activation timing of the stations constituting the multi-hop wireless system, determine the activation timer value,
In the third step,
The plant instrumentation monitoring control device determines whether or not to transmit the determined activation timer value to a station constituting the multi-hop wireless system based on a past activation timer value, and the determined activation timer If the value and the past start timer value are different, send the determined start timer value,
If the station receives the determined activation timer value, the activation timing of the station is controlled by the received activation timer value;
A power control method for a multi-hop wireless system. In the second step,
As the base point of the time of the activation timing, from the numerical calculation processing result obtained by the first step, extract the numerical calculation processing result at the time corresponding to a multiple of a predetermined period,
The numerical calculation processing results at the extracted periodic time points are compared with the boundary value of the allowable range set for each plant parameter in order from the time close to the time of the activation timing, and at the periodic time point within the allowable range. Find a multiple of that period,
Determining the next start timer value based on a multiple of that period;
The power control method for a multi-hop wireless system according to claim 1 or 2, characterized in that: In the third step,
The past activation timer value is the latest activation timer value transmitted;
The power control method for a multi-hop wireless system according to any one of claims 1 to 3, wherein: Numerical calculation processing for predicting the dynamic characteristics of the plant instrumentation process is performed by model predictive control using a step response model,
The power control method for a multi-hop wireless system according to any one of claims 1 to 4, wherein: Numerical calculation processing for predicting the dynamic characteristics of the plant instrumentation process is performed by model predictive control using an impulse response model,
The power control method for a multi-hop wireless system according to any one of claims 1 to 4, wherein: A plant instrumentation process for measuring plant parameters, a plant instrumentation supervisory control device for monitoring and controlling the plant instrumentation process, and a communication between the plant instrumentation process and the plant instrumentation supervisory control device. A multi-hop radio system composed of two or more stations that transmit in order, and a multi-hop radio system power control system comprising:
In the first step,
The multi-hop wireless system transmits a measurement signal converted by the plant instrumentation process by measuring a plant parameter to the plant instrumentation monitoring and control device,
The plant instrumentation monitoring and control device performs numerical calculation processing for predicting dynamic characteristics of the plant instrumentation process based on the received measurement signal,
In the second step,
The plant instrumentation monitoring and control device calculates a start timing of a station constituting the multi-hop wireless system based on the numerical calculation processing result, determines a start timer value,
In the third step,
The plant instrumentation monitoring control device determines whether or not to transmit the determined activation timer value to a station constituting the multi-hop wireless system based on a past activation timer value, and the determined activation timer Send value,
If the station receives the determined activation timer value, the activation timing of the station is controlled by the received activation timer value;
A power control system for a multi-hop wireless system. A plant instrumentation process for measuring plant parameters, a plant instrumentation supervisory control device for monitoring and controlling the plant instrumentation process, and a communication between the plant instrumentation process and the plant instrumentation supervisory control device. A multi-hop radio system composed of two or more stations that transmit in order, and a multi-hop radio system power control system comprising:
In the first step,
The multi-hop wireless system transmits a measurement signal converted by the plant instrumentation process by measuring a plant parameter to the plant instrumentation monitoring and control device,
The plant instrumentation monitoring and control device performs numerical calculation processing for predicting dynamic characteristics of the plant instrumentation process based on the received measurement signal,
In the second step,
The plant instrumentation monitoring and control device compares the numerical calculation processing result with a boundary value of a preset allowable range, and indicates a stable state period indicating that the numerical calculation processing result is within the allowable range And within the range of the estimated steady state period, calculate the activation timing of the stations constituting the multi-hop wireless system, determine the activation timer value,
In the third step,
The plant instrumentation monitoring control device determines whether or not to transmit the determined activation timer value to a station constituting the multi-hop wireless system based on a past activation timer value, and the determined activation timer If the value and the past start timer value are different, send the determined start timer value,
If the station receives the determined activation timer value, the activation timing of the station is controlled by the received activation timer value;
A power control system for a multi-hop wireless system. In the second step,
As the base point of the time of the activation timing, from the numerical calculation processing result obtained by the first step, extract the numerical calculation processing result at the time corresponding to a multiple of a predetermined period,
The numerical calculation processing results at the extracted periodic time points are compared with the boundary value of the allowable range set for each plant parameter in order from the time close to the time of the activation timing, and at the periodic time point within the allowable range. Find a multiple of that period,
Determining the next start timer value based on a multiple of that period;
The power control system for a multi-hop wireless system according to claim 7 or claim 8, wherein: In the third step,
The past activation timer value is the latest activation timer value transmitted;
The power control system for a multi-hop wireless system according to any one of claims 7 to 9, wherein: Numerical calculation processing for predicting the dynamic characteristics of the plant instrumentation process is performed by model predictive control using a step response model,
The power control system for a multi-hop wireless system according to any one of claims 7 to 10, wherein: Numerical calculation processing for predicting the dynamic characteristics of the plant instrumentation process is performed by model predictive control using an impulse response model,
The power control system for a multi-hop wireless system according to any one of claims 7 to 10, wherein:

Images