JP2022134764A - Sequence executing situation notification system and program - Google Patents

Abstract

To provide a sequence executing situation notification system and a sequence executing situation notification program capable of detecting a minute change in a facility where a frequency at which a congestion occurs is low.SOLUTION: A sequence executing situation notification system 21 includes an acquisition unit 32, a reference value storage unit 31, a difference degree calculation unit 33, and an image generation unit 36. The acquisition unit 32 acquires transition timing in response to a transition of a sequence step occurring during execution of a sequence. The acquisition unit 32 acquires as an operation timing time an elapsed time from the preceding transition timing immediately prior to an operation to the operation of an operated apparatus. The reference value storage unit 31 stores a reference value calculated based on a historical operation timing time. The difference degree calculation unit 33 identifies a reference value for an operation for which the operation timing time has been acquired, from the reference value storage unit 31 in response to acquisition of the operation timing time, and calculates a difference degree. The image generation unit 36 generates an image representing the difference degree and displays the image on a display unit 37.SELECTED DRAWING: Figure 3

Description

The present invention relates to a sequence execution status notification device and a sequence execution status notification program.

2. Description of the Related Art Many types of power generation equipment including hydroelectric power generation equipment equipped with water turbine generators and manufacturing equipment for petrochemical products are subjected to sequence control during starting and stopping. A sequence is executed by sequentially transitioning sequence steps, but there are cases where a so-called traffic jam occurs in which the transition is delayed and the sequence being executed is stopped.

Japanese Patent Laid-Open No. 2002-200002 discloses a sequence monitor device that detects an abnormality in a device that may potentially lead to a traffic jam before the traffic jam is detected. This sequence monitor device comprises a logic circuit, a timer circuit, a time counter, and an arithmetic unit, and monitors the operating state of each device element in each sequence step from start to stop of a device under sequence control. Detect traffic congestion. A logic circuit detects the state of the current sequence step, and an alarm signal is generated by the timer circuit if there is no transition to the next sequence step within a preset congestion monitoring set time. The time counter counts the actual operating time of the device element until transition to the next sequence step. The computing unit issues an alarm signal when the difference between the two predetermined integrated values is greater than the predetermined value, or when the slope of the function of the number of start/stops and the predetermined integrated value becomes greater than the predetermined value. .

JP-A-8-22311

However, as described above, the sequence monitor device of Patent Literature 1 issues an alarm signal when there is no transition to the next sequence step, so it cannot be said that it is necessarily suitable for equipment where the frequency of occurrence of traffic jams is low in the first place.

SUMMARY OF THE INVENTION Accordingly, it is an object of the present invention to provide a sequence execution status notification device and a sequence execution status notification program capable of detecting a slight change in facilities where traffic jams occur less frequently.

A sequence execution status notification device of the present invention includes an acquisition unit, a reference value storage unit, a difference calculation unit, and an image generation unit. The acquiring unit acquires the transition timing of the sequence step during execution of the sequence in which the sequence steps sequentially transition, identifies the immediately preceding transition timing from the already acquired transition timings as the preceding transition timing, and obtains the preceding transition timing. , the elapsed time until the operation of the operated device is acquired as the operation timing time. The reference value storage unit stores a reference value calculated based on a history operation timing time, which is a past operation timing time, for each operation of a plurality of devices that operated during the past execution of the sequence, in association with the operation of the device. do. The degree-of-difference calculation unit specifies the reference value of the motion whose motion timing time is acquired by the acquisition unit from the reference value storage unit in response to the acquisition of the motion timing time, and calculates the difference between the specified reference value and the motion timing time. Calculate degrees. The image generation unit generates an image representing the degree of difference and causes the display unit to display the image.

More preferably, the sequence execution status notification device includes a history information storage section and a reference value calculation section. The history information storage unit stores each operation of a plurality of devices that operated during the past execution of the sequence and the history operation timing time in association with each sequence. The reference value calculation unit calculates a reference value for each of the operations of the plurality of devices stored in the history information storage unit based on the history operation timing times stored in the history information storage unit.

More preferably, the sequence execution status notification device includes a specifying unit. The specifying unit is sequence-controlled, and specifies, in response to each operation of a device group including a plurality of devices, the operation associated with the historical operation timing time in the history information storage unit from among the operations of the operated device. and outputs the identified motion to the acquisition unit. The acquisition unit outputs the acquired operation timing time to the difference degree calculation unit in response to the input from the identification unit.

The reference value is associated with an abnormality probability indicating the probability that the operation of the device is abnormal, and the reference value calculation unit preferably calculates the reference value by learning the acquired history operation timing time.

It is preferable that the operation timing time acquired by the acquisition unit is used as a new historical operation timing time used for subsequent calculation of the reference value.

The sequence execution status notification program of the present invention causes a computer to execute an acquisition step, a reference value storage step, a difference calculation step, and an image generation step. The obtaining step obtains the transition timing of the sequence step during the execution of the sequence in which the sequence steps transition sequentially, identifies the immediately preceding transition timing from among the already obtained transition timings as the preceding transition timing, and obtains the preceding transition timing. , the elapsed time until the operation of the operated device is obtained as the operation timing time. In the reference value storage step, the reference value calculated based on the historical operation timing time, which is the past operation timing time, for each operation of the plurality of devices that operated during the past execution of the sequence, is stored in association with the operation of the device. Store in the value storage unit. In the difference calculating step, the reference value of the motion whose motion timing time is acquired by the acquiring step is specified from the reference value storage unit in response to the acquisition of the motion timing time, and the difference between the specified reference value and the motion timing time is determined. Calculate degrees. The image generating step generates an image representing the degree of difference and causes the display unit to display the image.

ADVANTAGE OF THE INVENTION According to this invention, a slight change in facilities with low frequency of congestion can be detected.

1 is a schematic diagram of a power generation system; FIG. FIG. 4 is an explanatory diagram of a startup sequence; It is a functional block diagram of an informing device. FIG. 10 is an explanatory diagram of calculation in a reference value calculation unit and a difference calculation unit; FIG. 10 is a graph of operation timing time with the start time of a sequence as a temporal starting point; FIG. It is a graph of operation timing time with the immediately preceding transition timing as a temporal starting point. FIG. 4 is an explanatory diagram of an image displayed on a display unit; FIG. 4 is an explanatory diagram of an image displayed on a display unit; FIG. 4 is an explanatory diagram of a transition timing period;

The power generation system 10 shown in FIG. 1 is one embodiment of the present invention. The power generation system 10 includes, for example, a power plant 11 provided in a power plant, and a sequence execution status reporting unit (hereinafter referred to as reporting unit) 12 electrically connected to the power plant 11 . The power plant 11 includes a power generation facility 15 , which is an example of equipment, and a controller 16 that causes the power generation facility 15 to execute a sequence, that is, performs control such as sequence control of the power generation facility 15 .

In this example, the power generation equipment 15 is equipment for performing hydroelectric power generation, and includes a water turbine generator (hereinafter simply referred to as "generator") 18A, an inlet valve 18B, a water turbine 18C, a power transmission section 18D, and the like. . The inlet valve 18B is connected to, for example, a water intake (not shown) of the dam and the water turbine 18C, and is opened and closed to switch between guiding water to the water turbine 18C and stopping guidance. The power transmission unit 18D sends the electricity generated by the generator 18A to the outside of a power plant (not shown) in which the power generation system 10 is provided, for example. The power generation equipment 15 has a group of devices in each part of the generator 18A, the inlet valve 18B, the water turbine 18C, the power transmission part 18D, and the connection parts of these parts. For example, the generator 18A has a control device (not shown) and an air cooler (not shown), the inlet valve 18B has an inlet valve main body (not shown), an opening/closing mechanism (not shown) for opening and closing the inlet valve main body, and the water turbine 18C. is equipped with runners (not shown) and guide vanes (not shown), and the power transmission section 18D is equipped with devices such as a parallel circuit breaker (not shown). In FIG. 1, in order to avoid complication of the drawing, only the generator 18A is shown as equipment, and only five equipments 19a to 19e are shown.

The controller 16 makes the power generation equipment 15 execute a sequence by controlling each part of the power generation equipment 15 in an integrated manner. The controller 16 performs control related to the operation (including stopping) of the power generation equipment 15, and sequence control is one of the controls related to the operation. In some cases, the controller 16 acquires an operation signal indicating the operation of each device in the device group from the power generation facility 15 and performs feedback control on the power generation facility 15 based on the acquired operation signal. An operation signal is generated for each operation in the power generation equipment 15 and output to the controller 16 . The motion signal includes a motion signal that is binary information about motion and a motion signal that is not binary information. Binary information about operation is a signal that switches between one and the other of two possible states of a device, for example, a signal that switches between on and off, a signal that switches between 0 (zero) and 1, and the opening and closing of a valve, etc. is a signal that switches between an open state and a closed state. Operation signals that are not binary information can be obtained, for example, in the power generation equipment 15, such as the level of the lubricating oil level in which the bearings of the rotating shaft of the water turbine 18C are immersed, the temperature of the lubricating oil, and the like. It is a signal indicating a value according to a state, such as a detection signal when values are continuous.

The controller 16 of this example transmits all of the operation signals, which are binary information, among the operation signals indicating the operation of the equipment operated in the power generation facility 15 during execution of the sequence, to the sequence execution status notification device (hereinafter referred to as the notification unit) of the notification unit 12. , annunciation device) 21. The controller 16 further outputs to the notification device 21 a transition signal generated in the power generation facility 15 and indicating that the sequence has transitioned. The controller 16 may regard a specific operation signal, which is a binary signal related to operation generated in the power generation facility 15 , as a sequence transition signal and output the transition signal to the notification device 21 . In the above example, the controller 16 specifies the motion signal, which is binary information, from the acquired motion signal, and outputs the specified motion signal to the notification device 21 . However, the operation signal, which is binary information, may be directly output from each device of the power generation facility 15 to the notification device 21 without going through the controller 16 . In this example, when the device performs switching operation from one side to the other side, that is, when the controller 16 acquires an operation signal that is binary information from the power generation facility 15, the lighting state changes from non-lighting to non-lighting or from lighting to non-lighting. The controller 16 is provided with an indicator light that changes (switches) to lighting. An operation signal is output to the notification device 21 when the lighting state of the indicator lamp changes from the start of the sequence. Further, in this example, the controller 16 is provided with an indicator lamp that similarly changes (switches) when a transition signal is input. 21 is output.

The power generation system 10 of this example includes a notification unit 12 composed of a notification device 21 and a plurality of terminals 22a to 22c, but the plurality of terminals 22a to 22c may be omitted. That is, a power generation system configured by the power generation plant 11 and the notification device 21 may be used. In the following description, the terminals 22a to 22c will be referred to as terminals 22 when they are not distinguished from each other.

The notification device 21 notifies the execution status of the sequence being executed in the power generation equipment 15 based on the transition signal and the operation signal input from the power plant 11 (in this example, the transition signal and the operation signal input from the controller 16). do. Specifically, when the power generation equipment 15 is activated, the probability of congestion, that is, the possibility of activation congestion, is reported as the execution status of the sequence. Note that the notification device 21 can also notify the probability of a traffic jam (stop traffic jam) occurring when the power generation equipment 15 is stopped, as the execution status of the sequence.

The notification device 21 and each of the plurality of terminals 22a to 22c in this example are configured to be able to communicate with each other, and the notification device 21 can also notify the terminal 22 of the execution status of the sequence by transmitting it to the terminal 22. It's like Transmission from the notification device 21 to the terminal 22 is performed, for example, in response to the acquisition of a transmission instruction from the terminal 22 to the notification device 21 by the notification device 21 . The terminal 22 in this example is, for example, a personal computer (hereinafter referred to as PC) having a display unit (not shown) such as a liquid crystal display. ) and the like. Note that the number of terminals 22 provided in the notification unit 12 is not limited to three in this example.

A sequence is executed by sequentially transitioning a plurality of sequence steps. In the following description, a plurality of ordered sequence steps will be referred to as sequence steps S0 (starting sequence step), S1, S2, . . . A plurality of sequence steps S for executing a sequence and their number (the number of steps) differ depending on the power plant. For example, in the case of a start-up sequence (hereinafter referred to as a start-up sequence), different power plants A to C execute sequences with different sequence steps S0, S1, S2, . . . as shown in Table 1. Also, the number of steps in the sequence step S is different from each other. The same applies to the stop sequence (hereinafter referred to as stop sequence).

In Table 1, "preparation" is a sequence that prepares the conditions necessary for starting the generator 18A, such as full closure of the inlet valve 18B and guide vanes (not shown) and opening of the parallel circuit breaker (not shown). Step S. "Inlet valve" is a sequence step S for fully opening the inlet valve 18B (see FIG. 1). "Start" is a sequence step S in which the water turbine 18C is started by gradually increasing the opening of the guide vanes (not shown) by a starting device (not shown) on the condition that the inlet valve 18B is fully open. "Excitation" is a sequence step S in which the generator 18A is excited by an exciter (not shown). "Parallel" is a sequence step S for connecting the power generation equipment 15 to the system. In the case of the stop sequence, there are a sequence step S of "disconnection" for disconnecting the power generation equipment 15 from the system and a sequence step S of "fully closed inlet valve" for fully closing the inlet valve.

In the startup sequence of the power plant 11, the transition timing to transition to sequence step S0 is assumed to be transition timing TT0. Similarly, transition timings for transitioning to sequence steps S1, S2, . . . are assumed to be transition timings TT1, TT2, . In the power plant 11, as shown in FIG. 2, when the time axis is the horizontal axis and the passage of time is shown rightward on the paper surface, the transition timing TT0 at which the transition to the sequence step S0 is made is the starting point, and the transition is made rightward. At timings TT1, TT2, TT3, TT4 and TT5, sequence step S transitions to sequence steps S1, S2, S3, S4 and S5 in order. This causes the boot sequence to be executed. In the following description, the transition timings TT0, TT1, TT2, .

3, the notification device 21 includes a reference value storage unit 31, an acquisition unit 32, a difference calculation unit 33, and an image generation unit . The notification device 21 may have a display section 37, which is also the case in this example. Preferably, the notification device 21 further includes a history information storage unit 41, a reference value calculation unit 42, and a specification unit 43, as in this example.

The acquisition unit 32 is electrically connected to the controller 16 and receives the transition signal output from the controller 16 . The acquisition unit 32 has a timer 46 that counts the elapsed time from the transition timing TT0, and turns on the counting by the timer 46 in response to the input of the transition signal of the sequence step S0 from the controller 16. FIG. Also, the obtaining unit 32 obtains the timer values counted by the timer 46 as the transition timings TT1, TT2, . Thus, the acquisition unit 32 acquires the transition timings TT0, TT1, TT2, . . . in response to the transition of the sequence step S during execution of the sequence. Note that in this example, the transition signal is generated in the power plant 11 (see FIG. 1), and the already generated transition signal is input to the acquisition unit 32 . However, the acquisition unit 32 may generate the transition signal by identifying a predetermined signal input from the power plant 11 as the transition signal.

An acquisition unit 32 connected to the controller 16 receives an operation signal output from the controller 16 . The input motion signal is a signal of binary information about motion. In response to the input of the operation signal, the obtaining unit 32 reads out the timer value counted by the timer 46 as the operation timing of the operation indicated by the operation signal, and selects the transition timing TT immediately before the operation timing from among the already obtained transition timings TT. is specified as the preceding transition timing. Then, by subtracting the timer value of the preceding transition timing from the timer value of the operation timing, the operation timing time, which is the elapsed time from the transition timing immediately before the operation to the operation timing, is calculated and obtained. In this way, in response to the operation of the device, the acquisition unit 32 identifies the preceding transition timing from among the already acquired transition timings TT, and operates the elapsed time from the preceding transition timing to the operation of the device that has operated. Get it as a timing time. Note that when the operation signal, which is binary information output from the power generation equipment 15 , is input to the acquisition unit 32 without going through the controller 16 , the acquisition unit 32 may be electrically connected to the power generation equipment 15 . Also, in this example, an operation signal of binary information is generated in the power plant 11 (see FIG. 1), and the already generated operation signal is input to the acquisition unit 32 . However, the acquisition unit 32 may generate an operation signal of binary information by specifying a signal of predetermined binary information input from the power plant 11 as an operation signal.

The notification device 21 of this example includes an identification unit 43 , and the acquisition unit 32 outputs to the identification unit 43 an operation signal indicating the operation for which the operation timing time is acquired. When the motion signal output to the specifying unit 43 is input from the specifying unit 43 as described below, the acquisition unit 43 acquires timing time information that associates the motion indicated by the input motion signal with the motion timing time of the motion. are output to the difference calculation unit 33 and the history information storage unit 41, respectively. The output of the acquisition unit 32 to the history information storage unit 41 associates the motion and the motion timing time with the sequence indicating the preceding transition timing, and stores the history information with the motion as the history motion and the motion timing time as the history motion timing time. This is done by storing in the unit 41 . In this manner, the acquisition unit 32 generates history information in which history actions, history action timing times, and sequences are associated with each other, and stores the history information in the history information storage unit 41 . When the storage in the history information storage unit 41 is completed, the acquisition unit 32 also notifies the reference value calculation unit 42 of the completion of storage in the history information storage unit 41 (storage completion notification).

The identification unit 43 is connected to the acquisition unit 32 , and in response to the input of the operation signal from the acquisition unit 32 , stores the same past operation as the operation indicated by the input operation signal in the history operation of the history information storage unit 41 . identify from within. If it is identified, that is, if it is determined that there is a motion that is the same as the motion indicated by the input motion signal in the history motions associated with the history motion timing time, the acquisition unit 32 acquires the motion. Outputs an operation signal indicating If it is not identified, that is, if it is determined that there is no motion that is the same as the motion indicated by the input motion signal among the history motions associated with the history motion timing time, the acquisition unit 32 is operated. No signal output. Note that instead of making a determination based on whether or not the history motion is specified, the same motion as the motion indicated by the input motion signal is included in the history information of a predetermined percentage or more, such as 90%, for example. It may be judged that the

In this example, the specifying unit 43 is connected to the acquiring unit 32 and an operation signal is input to the specifying unit 43 via the acquiring unit 32 . However, instead of this aspect, the specifying unit 43 may be connected to the controller 16 and an operation signal, which is binary information about the operation, may be input from the controller 16 to the specifying unit 43 . In this case, the identification unit 43 may also be provided with a timer similar to the timer 46 , the operation timing time may be obtained by the identification unit 43 , and the obtained operation timing time may be output to the acquisition unit 32 . However, from the viewpoint of increasing the accuracy of the operation timing time, it is more preferable that the timer value for obtaining the transition timing and the timer value for the operation timing for obtaining the operation timing time are counted by a common timer.

The history information storage unit 41 associates, for each sequence, the operation of the device that operated during the past execution of the sequence (hereinafter referred to as a history operation) with the operation timing time of the history operation. Stored as history information (see Table 2). When the acquiring unit 32 newly acquires the operation timing time by a newly executed sequence, the acquired new operation timing time and the operation are stored in the history information storage unit 41 as the history operation timing time and the history operation. is stored as Since one sequence has a plurality of operations and history information is generated for each history operation, a plurality of pieces of history information are generated and stored by executing the sequence once. Note that if the historical motion and the historical motion timing time are obtained by other acquiring means or the like different from the acquiring unit 32, the obtained historical motion and the historical motion timing time are stored in advance in the history information storage unit 41, for example. may be stored in

In response to acquisition of the storage completion notification from the acquisition unit 32, the reference value calculation unit 42 calculates the history operation timing time associated with the stored history operation and the past history operation timing time of the stored history operation. are identified, and a reference value is calculated based on their historical operation timing times. In this manner, the reference value calculation unit 42 calculates the reference value for each history operation based on the history operation timing time stored in the history information storage unit 41 . The calculation method of the reference value is not limited as long as it is a value statistically derived from the history operation timing time. The reference value is a value associated with an abnormality probability that indicates the probability that the operation of the device is abnormal.

The reference value calculation unit 42 is preferably a learning calculation unit that learns the acquired history operation timing time and calculates the reference value. In this example, the learning calculation unit is used as the reference value calculation unit 42 as well. The reference value calculation unit 42 of this example creates a distribution of the history operation timing times using a non-parametric method from the history operation timing times from which the outliers are removed, and calculates the reference value from the statistical information of the data. ing. Although it is not necessary to remove the outliers, it is preferable to remove the outliers because the probability of the possibility of start-up congestion can be reported more accurately. Outlier removal uses One-Class SVM twice to remove outliers. It is preferable to remove excessively large outliers in historical operation timing times with a first application of the One-Class SVM, and remove small, subtle outliers with a second application of the One-Class SVM.

The historical operation timing times hM from which outliers are removed are used as learning data, and the distribution of the historical operation timing times is obtained from the historical operation timing times hM using kernel density estimation, which is a non-parametric density estimation method (( A)). The obtained distribution is the distribution of history operation timing times in a state of normal operation. From this distribution, the elapsed time α1, which is highly likely to be frequently observed at normal history operation timing _times , and the elapsed time α2 slightly _deviating from α1 _are obtained. For example, α1 is the elapsed _time to which 65% of the historical operation timing time distribution belongs, and α2 is obtained by multiplying the standard deviation s of the historical operation timing time distribution by _a preset coefficient and adding it to _α1 . Also in this example, it is obtained as a value obtained by adding ₁₂₀ ×(the standard deviation s of the distribution of the history operation timing time) to α1. α ₁ may be the elapsed time to which the history operation timing time distribution belongs at a predetermined rate, and is not limited to 65% as in the present example, but is set to at least 50%. It is more preferable to ensure the validity of the probability.

In this example, the _first reference value VA (see _FIG . 4 ) and a second reference value VB (see FIG. 4) associated with an abnormality probability higher than the first reference value VA are calculated by the following equations (1) and (2), respectively. Then, the abnormality probability y1 obtained with the _first reference value VA is set at 2%, for example, and the abnormality probability y2 obtained with the _second reference value VB is set at 80%, for example. The abnormality probability y ₁ obtained with the first reference value VA and the abnormality probability y ₂ obtained with the second reference value are not limited to this example. The abnormality probability to be taken for each reference value may be appropriately set based on the performance of the probability model to be generated. For example, if you want to calculate the probability with a margin around the reference value, set the abnormality probability to a low value, and if you want to calculate the probability strictly around the reference value, set the abnormality probability to a high value. good.
VA=α ₁ (1)
VB={(α ₂ −α ₁ )/s}+α ₁ (2)

The reference value storage unit 31 associates the reference value with the history operation and stores them as reference value information (see Table 3).

In response to the input of the timing time information including the operation timing time acquired by the acquisition unit 32, the difference calculation unit 33 stores the same operation as the operation included in the timing time information among the history operations of the reference value storage unit 31. , and the reference value associated with that historical behavior. The difference calculation unit 33 calculates the difference between the acquired operation timing time and the reference value specified from the reference value storage unit 31 , and outputs the difference to the image generation unit 36 .

The dissimilarity is obtained using the following method. First, each of the abnormality probability y ₁ obtained by the first reference value VA and the first reference value VA and the abnormality probability y ₂ obtained by the second reference value VB and the second reference value VB is converted into the sigmoid function of the following equation (3). Constants A and B are calculated by matching them. Then, for the abnormality probability model PM generated from the sigmoid function of the following equation (3) using the calculation constants A and B, the abnormality probability y (where 0.0 ≤ y ≤ 1 .0) is calculated. That is, by obtaining constants A and B, a probability model is generated.
y=1/(1+ ^eAx+B ) (3)

The obtained probability model is graphed (see (B) of FIG. 4), and the horizontal axis of this graph ((B) of FIG. 4) indicates the operation timing time MTn of the acquired operation Mn as x By taking it as a value (see (C) in FIG. 4), the abnormality probability Pn of the acquired motion Mn can be represented on the vertical axis as y in the above equation (3). Since the displayed abnormality probability Pn shows a difference from the abnormality probability in the reference value, this abnormality probability can be used as the dissimilarity. In this example, the abnormality probability is expressed as a percentage with a unit of %.

The image generation unit 36 generates an image representing the degree of dissimilarity calculated by the degree of dissimilarity calculation unit 33 and outputs image data representing the image to the display unit 37 to display the image on the display unit 37 . The display unit 37 is, for example, a liquid crystal display.

The notification device 21 is composed of a computer, and by causing the computer to execute a predetermined program, the computer functions as each part described above. The program is a sequence execution status notification program that causes a computer to execute an acquisition step, a reference value storage step, a difference calculation step, and an image generation step. The obtaining step obtains the transition timing in response to the transition of the sequence step S during the execution of the sequence in which the sequence step S transitions sequentially, and obtains the transition timing from the already obtained transition timing in response to the operation of the device. The immediately preceding transition timing is specified as the preceding transition timing, and the elapsed time from the preceding transition timing to the operation of the device that has operated is acquired as the operation timing time. The reference value storage step associates the reference value calculated based on the historical operation timing time, which is the past operation timing time, for each operation of the plurality of devices that operated during the past execution of the sequence, with the operation of the device ( As reference value information/associated reference value information) is stored in the reference value storage unit. In the difference calculating step, the reference value of the motion whose motion timing time is acquired by the acquiring step is specified from the reference value storage unit in response to the acquisition of the motion timing time, and the difference between the specified reference value and the motion timing time is determined. Calculate degrees. The image generating step generates an image representing the degree of difference and causes the display unit to display the image.

The operation of the above configuration will be explained. In the power generation plant 11 of the power generation system 10 , the power generation facility 15 is activated, generated, and stopped under the control of the controller 16 . The power generation equipment 15 executes a sequence by the controller 16 to start up, and performs power generation and power transmission. Similarly, the sequence is executed by the controller 16 and stopped. The controller 16 transmits the operation signal of the binary information about the operation among the operation signals generated by the power generation equipment 15 to the acquisition unit 32 of the notification device 21 regarding the operation of the equipment operated by the power generation equipment 15 during execution of the sequence. Output.

When execution of the startup sequence is started, the controller 16 outputs a transition signal to the acquisition unit 32 indicating that the process has transitioned to the sequence step S0. The controller 16 then outputs the generated operation signals and transition signals to the acquisition unit 32 in the order in which they were generated. When acquiring the transition signal indicating the transition to the sequence step S0, the acquisition unit 32 turns on the start of counting by the timer 46 and stores the transition timing TT0 in the storage unit built in the notification device 21 (in this example, the acquisition unit 32 It is stored in a provided storage unit (not shown) in association with the sequence step S0. After that, when a transition signal is input, the acquisition unit 32 acquires the transition timings TT1, TT2, . After obtaining the subsequent transition timing TT, the sequence period as the time between the transition timings is obtained by subtracting the timer value indicating the preceding transition timing TT from the timer value indicating the subsequent transition timing TT. This sequence period may also be determined and stored in the storage unit (not shown) in association with the transition signal.

When the transition signal and the operation signal are respectively input, the obtaining unit 32 identifies the immediately preceding transition timing TT from among the transition timings stored in the storage unit, and obtains the operation timing time based on the timer value. . When the operation signal is input, the acquisition unit 32 also outputs the operation signal to the identification unit 43 .

In response to the input of the motion signal from the acquisition unit 32 , the specifying unit 43 specifies the same past motion as the motion indicated by the input motion signal from among the history motions of the history information storage unit 41 . When specified, an operation signal indicating the operation is output to the acquisition unit 32 . If not specified, no operation signal is output to the acquisition unit 32 .

For example, when a motion signal indicating motion M1 is input to the specifying unit 43, the history motion indicating M1 in the history information storage unit 41 shown in Table 2 is the history information A of sequence A, the history information B of sequence B, and the sequence C. In this case, the identification unit 43 outputs to the acquisition unit 32 a motion signal indicating the motion M1. On the other hand, when the motion signal indicating the motion M3 is input to the specifying unit 43, the historical motion indicating the motion M3 in the history information storage unit 41 shown in Table 2 is not specified in any of the history information A to C. Therefore, in this case, the operation signal indicating the operation M3 is not output to the acquisition unit 32 .

The history information storage unit 41 stores motions related to the past execution of the sequence as history motions, and the specifying unit 43 stores the same motion as the motion indicated by the motion signal input via the acquisition unit 32 in the history. It is specified from among the history operations of the information storage unit 41 . As a result, the operation signal indicating the operation of the equipment operating during the execution of the sequence is extracted from the operation signals input from the power plant 11 to the acquisition unit 32, and the extracted operation signal and the operation indicated by the operation signal are extracted. are output to the difference degree calculation unit 33 and the history information storage unit 41, respectively. Therefore, only the extracted motion signal is processed by the dissimilarity calculation unit 33, so the processing efficiency up to notification is better. Further, since the extracted motion signal and the motion timing time of the motion indicated by the motion signal are stored in the history information storage unit 41 as a new history motion and history motion timing time, the probability of subsequent extraction is increased. Along with the improvement, the accuracy of the calculation result by the reference value calculation unit 42 using the history operation timing time is also improved.

When the motion signal is input from the specifying unit 43 , the acquisition unit 32 outputs timing time information that associates the acquired motion timing time with the motion to the difference degree calculation unit 33 and the history information storage unit 41 . . When the operation timing time is the elapsed time from the start of the sequence (transition timing TT0 in this example), when focusing on a specific operation signal, the difference between the sequences is as shown in FIG. It is small and it is difficult to understand the size of the difference between them. Therefore, it is difficult to notice that it is abnormal. In addition, when different operation signals are compared, it is difficult to understand the difference between the operation signals, and it is difficult to determine whether or not the operation signal is abnormal, so that it should be noted. Therefore, the certainty of the value for each operation timing time is low. On the other hand, in this example, since the operation timing time is the elapsed time from the previous transition timing, when focusing on a specific operation signal, as shown in FIG. are displayed, and the magnitude of the mutual difference is easy to understand. Therefore, it is easy to notice that something is abnormal or the possibility of being abnormal. In addition, when different operation signals are compared with each other, the difference between the operation signals is conspicuous, and it is easy to determine whether or not the operation signal is abnormal, which should be noted. Therefore, the likelihood of the value for each operation timing time is high, and the abnormality probability can be obtained with higher accuracy. Therefore, even a slight change in the power generation equipment 15 where traffic jams occur less frequently is detected. 5 and 6 are the motions M1, M2, M3, . . . , and in FIGS. The number of operations belonging to sequence step S is not limited to the number shown in FIG. Also, the vertical axes of (B) to (E) in FIG. 6 are history operation timing times (in seconds) as in (A), but they are omitted in FIG.

When the timing time information including the operation timing time is input, the dissimilarity calculation unit 33 selects the same operation as the operation included in the timing time information from the history operation of the reference value storage unit 31 in response to the input. Identify and specify reference values. For example, it is assumed that the timing time information including the action M1 is input to the dissimilarity calculator 33, and reference value information IV1, IV2, and IV5 are stored in the reference value storage unit 31 as shown in Table 3. Reference value information IV1 includes history operation M1, first reference value VA1 and second reference value VB1, reference value information IV2 includes history operation M2, first reference value VA2 and second reference value VB2, It is assumed that reference value information IV5 includes history operation M5, first reference value VA5, and second reference value VB5. In this case, among the reference value information IV1, IV2, and IV5, the historical operation indicating M1 is included in the reference value information IV1. and the second reference value VB2. The difference calculation unit 33 calculates the difference between the operation timing time input from the acquisition unit 32 and the reference value. In the above example, both the input operation timing time and the first reference value VA1 and the second reference value VB1 are used as reference values to obtain the degree of difference. value. As described above, the operation timing and the reference value used for the calculation are the values of the time that is not affected by the time before the immediately preceding transition timing TT, so the degree of difference obtained has a high degree of certainty.

When the acquisition unit 32 outputs history information in which the timing time information is associated with the sequence, the operation is the history operation, and the operation timing time is the history operation timing time, to the history information storage unit 41, the reference value calculation unit 42 calculates the reference value. calculate. That is, when new history information is stored in the history information storage unit 41, the reference value calculator 42 calculates the reference value for the history action included in the new history information.

For example, in the above-described case shown in Table 2, assume that the acquisition unit 32 outputs the operation M1 as a history operation and the operation timing time hM1 as a history operation timing time to the history information storage unit 41 in association with the sequence D. In the history information storage unit 41 in this case, the history operation M1 is specified for each of the sequence A, the sequence B, and the sequence C, and the reference value calculation unit 42 associates each of the history operations M1 of these sequences A to C. A new reference value is calculated from one history operation timing time hM1 and the newly stored history operation timing time hM1 of the history operation M1 of the sequence D. Further, in the above-described case shown in Table 2, assume that the acquisition unit 32 sets the operation M3 as the history operation and the operation timing time hM3 as the history operation timing time, and associates them with the sequence D and outputs them to the history information storage unit 41 . In this case, in the history information storage unit 41, since the history operation M3 is not identified in any of sequence A, sequence B, or sequence C, the reference value calculation unit 42 determines the history operation M3 of the newly stored sequence D. A new reference value for the operation M3 is calculated only from the historical operation timing time hM3.

In this manner, the reference value calculation unit 42 calculates the reference value for each of the history actions based on the history action timing times stored in the history information storage unit 41 . Since the history operation timing time is the elapsed time from the immediately preceding transition timing TT, it does not include the influence of the elapsed time before the immediately preceding transition timing TT. Therefore, the reference value statistically obtained from the history operation timing time is also a value that does not include the influence of the elapsed time before the immediately preceding transition timing TT.

The reference value calculation unit 42 is a learning calculation unit that learns the acquired history operation timing time and calculates the reference value. is high.

When the degree of difference is input, the image generation unit 36 generates an image representing the degree of difference, and causes the display unit 37 to display the image. The abnormality probability is notified by displaying an image of the degree of difference, which is the degree of difference from the reference value, on the display unit 37 . For example, as shown in FIG. 7, the display unit 37 generates an image G1 including the abnormality probability Pn, which is the degree of difference between the operation timings MTn of the operations Mn (n is a natural number), and displays the image G1 on the display unit 37 (see FIG. 2). Then, the abnormality probability Pn of the obtained motion Mn is reported. In the example shown in FIG. 7, the obtained motion Mn is plotted on the graph of the abnormality probability model PM shown in FIG. Although it is an example, the image to be displayed is not limited to this example. For example, only the numerical value of the obtained abnormality probability Pn may be displayed as the image G1, or in addition to or instead of the numerical value, the abnormality probability Pn may be indicated by color contrast or color.

Each time an image G1 representing the degree of difference is generated, the motions for which the degree of difference has been calculated are added to display a list of the motions for which the degree of difference has been calculated. When a selection input is made, the degree of difference of the selected action may be displayed on the display unit 37 in response to the input. For example, the image generator 36 may generate an image G2 displaying the sequence execution status together with the transition status of the sequence step S, as shown in FIG. In the image G2, each operation timing of the operation Mn (in FIG. 8, n is 1, 2, 4, 5, 7, 8) is associated with the preceding sequence step S and displayed as a list. is displayed as the abnormality probability Pn. The display unit 37 is connected to a mouse (not shown) as an input unit (not shown), and by operating the mouse, a cursor CU is placed on a predetermined column CL and clicked to select the column CL. . In response to this selection operation, when a selection notification indicating that the column CL has been selected is input to the image generation unit 36, in response to this input, the image generation unit 36 converts the image G2 into the selected column CL. to an image G1 (see FIG. 7) showing the degree of difference (abnormality probability Pn) for the motion Mn. The selection of the column CL is not limited to the mouse, and may be performed by using, for example, a keyboard as the input unit or a touch panel display as the display unit 37 . Further, the image generator 36 may include a switch button or the like for switching to the image G2 in the image G1 (see FIG. 7). may

As described above, since an image of the degree of difference is displayed for each of the motion signals of the binary information about the motion, it is easy to determine the individual motions related to the sequence and the potential abnormality of the device that is the subject of the motion. Furthermore, according to the above configuration, since the degree of difference is displayed for each operation related to the sequence, congestion occurs before the transition timing to the sequence step (succeeding sequence step) S after the operation signal is obtained. Actions that reduce probability and cause congestion are reliably identified. Further, as described above, the terminal 22 in this example has a display section, and an image representing the degree of difference is also displayed on the display section of the terminal 22 . Therefore, the terminal 22 also displays the abnormality probability.

In the above example, the temporal location (position on the time axis) within the sequence of motion indicated by the motion signal is specified using the transition timing TT as an index. As for the operation at a timing close to the transition timing TT to the succeeding sequence step S, the operation signal may be obtained after the transition timing TT of the succeeding sequence step S. In the case of the operation signal for such an operation, when the immediately preceding transition timing TT is specified as the transition timing TT to the preceding sequence step S, and when it is specified as the transition timing TT to the subsequent sequence step S and both. Therefore, for each sequence step S, it is preferable to include, for example, a predetermined surplus time in the transition timing TT to the succeeding step timing S in the duration of the preceding sequence step S. That is, as shown in FIG. 9, the transition timing periods SP0, SP1, SP2, .・It is preferable to consider As a result, an operation signal for an operation whose operation timing is close to the transition timing to the succeeding sequence step S can be more reliably obtained at the operation timing time that has elapsed from the transition timing of the preceding sequence step S. When the operation timing is within the surplus time Tm, the preceding sequence step S including the surplus time in the transition timing period SPn (n is a natural number equal to or greater than 1) is associated with the transition timing period SPn+1. It is preferred to identify both the sequence step S of a row as the immediately preceding sequence step S. As a result, the abnormality probability can be obtained more reliably for each motion indicated by the motion signal.

The surplus times Tm of the transition timing periods SP0, SP1, SP2, . . . may be the same or different. If the approximate time length between the transition timings TT is known from past data or can be predicted, the transition timing period SPn (n is a natural number of 1 or more) of the preceding step timing ) is preferably set to a time shorter than the transition timing period SPn+1 of the subsequent step timing. As a result, the transition timing immediately before the operation indicated by the operation signal is specified as the transition timing of at most two sequence steps S, that is, the transition timing TT of the preceding sequence step S and the transition timing TT of the following sequence step S. This prevents the transition timing TT of the sequence step S next to the succeeding sequence step S from being specified. Therefore, the abnormality probability can be obtained more reliably for each motion indicated by the motion signal.

In this example, the surplus time Tm included in the transition timing period of the final sequence step S is set to 60 seconds as in sequence step 5 shown in FIG. seconds). The surplus time Tm included in the transition timing period of the final sequence step S is preferably in the range of 30 seconds to 180 seconds, and the surplus time Tm included in the other transition timing periods is set at 20 seconds at the longest. It is preferable to set the time to 10 seconds at the longest.

In this embodiment, the power generation equipment 15 including the water turbine generator is used as the target of the sequence control, that is, the equipment on which the sequence is executed. However, the equipment is not limited to this, and may be other power generation equipment, plant equipment, or the like, as long as the equipment executes the sequence.

10 power generation system 11 power plant 12 sequence execution status notification unit 15 power generation equipment 16 controller 18A generator 18B inlet valve 18C water wheel 18D power transmission unit 21 sequence execution status notification device 22a to 22c terminal 31 reference value storage unit 32 acquisition unit 33 difference calculation Section 36 Image Generation Section 37 Display Section 41 History Information Storage Section 42 Reference Value Calculation Section 43 Identification Section 46 Timer

A sequence execution status notification device of the present invention includes an acquisition unit, a reference value storage unit, a difference calculation unit, and an image generation unit. The acquisition unit acquires a transition timing at which a sequence step transitions to the next sequence step in response to the transition of the sequence step during the execution of a sequence in which the sequence step transitions sequentially in a facility having a plurality of devices, and a plurality of In response to each action of the device, the transition timing immediately before the action of the device that has acted is specified as the preceding transition timing from among the transition timings that have already been acquired, and the progress from the preceding transition timing to the action of the device that has acted Get the time as the operation timing time. The reference value storage unit stores a reference value calculated based on a history operation timing time, which is a past operation timing time, for each operation of a plurality of devices that operated during the past execution of the sequence, in association with the operation of the device. do. The degree-of-difference calculation unit specifies the reference value of the motion whose motion timing time is acquired by the acquisition unit from the reference value storage unit in response to the acquisition of the motion timing time, and calculates the difference between the specified reference value and the motion timing time. Calculate degrees. The image generation unit generates an image representing the degree of difference and causes the display unit to display the image.

The sequence execution status notification program of the present invention causes a computer to execute an acquisition step, a reference value storage step, a difference calculation step, and an image generation step. The acquisition step acquires the transition timing at which the sequence step transitions to the next sequence step in response to the transition of the sequence step during the execution of the sequence in which the sequence steps sequentially transition in the equipment having a plurality of devices, and a plurality of In response to each action of the device, the transition timing immediately before the action of the device that has acted is specified as the preceding transition timing from among the transition timings that have already been acquired, and the progress from the preceding transition timing to the action of the device that has acted Get the time as the operation timing time. In the reference value storage step, the reference value calculated based on the historical operation timing time, which is the past operation timing time, for each operation of the plurality of devices that operated during the past execution of the sequence, is stored in association with the operation of the device. Store in the value storage unit. In the difference calculating step, the reference value of the motion whose motion timing time is acquired by the acquiring step is specified from the reference value storage unit in response to the acquisition of the motion timing time, and the difference between the specified reference value and the motion timing time is determined. Calculate degrees. The image generating step generates an image representing the degree of difference and causes the display unit to display the image.

The notification device 21 of this example includes an identification unit 43 , and the acquisition unit 32 outputs to the identification unit 43 an operation signal indicating the operation for which the operation timing time is acquired. When the motion signal output to the specifying unit 43 is input from the specifying unit 43 as described below, the acquisition unit 32 acquires timing time information that associates the motion indicated by the input motion signal with the motion timing time of the motion. are output to the difference calculation unit 33 and the history information storage unit 41, respectively. The output of the acquisition unit 32 to the history information storage unit 41 associates the motion and the motion timing time with the sequence indicating the preceding transition timing, and stores the history information with the motion as the history motion and the motion timing time as the history motion timing time. This is done by storing in the unit 41 . In this manner, the acquisition unit 32 generates history information in which history actions, history action timing times, and sequences are associated with each other, and stores the history information in the history information storage unit 41 . When the storage in the history information storage unit 41 is completed, the acquisition unit 32 also notifies the reference value calculation unit 42 of the completion of storage in the history information storage unit 41 (storage completion notification).

Claims (6)

During execution of a sequence in which sequence steps transition sequentially, the transition timing of the sequence step is acquired, and the preceding transition timing is specified as the preceding transition timing from among the already acquired transition timings, and the preceding transition timing is determined from the preceding transition timing. , an acquisition unit that acquires the elapsed time until the operation of the operated device as the operation timing time;
A reference value storage unit that stores a reference value calculated based on a history operation timing time, which is a past operation timing time, for each operation of a plurality of devices that operated during the past execution of the sequence, in association with the operation of the device. When,
specifying the reference value of the motion for which the operation timing time is acquired by the acquisition unit from the reference value storage unit in response to acquisition of the operation timing time, and determining the specified reference value and the operation timing time; a dissimilarity calculation unit that calculates the dissimilarity of
A sequence execution status notification device, comprising: an image generation unit that generates an image representing the degree of difference and displays it on a display unit. a history information storage unit that stores each operation of a plurality of devices that operated during the past execution of the sequence and the history operation timing time in association with each sequence;
a reference value calculation unit that calculates the reference value based on the history operation timing time stored in the history information storage unit for each of the operations of the plurality of devices stored in the history information storage unit;
The sequence execution status notification device according to claim 1, comprising: The operation associated with the history operation timing time in the history information storage unit is specified from among the operations of the equipment that has been sequence-controlled in response to the operation of each of the equipment group including the plurality of equipment. , further comprising a specifying unit that outputs the specified operation to the acquiring unit,
3. The sequence execution status notification device according to claim 2, wherein the acquisition unit outputs the acquired operation timing time to the difference degree calculation unit in response to the input from the identification unit. The reference value is associated with an abnormality probability indicating the probability that the operation of the device is abnormal,
4. The sequence execution status notification device according to claim 2, wherein the reference value calculation unit learns the acquired history operation timing time to calculate the reference value. 5. The sequence execution status notification according to any one of claims 1 to 4, wherein the operation timing time acquired by the acquisition unit is set as the new history operation timing time used for calculating the reference value from the next time onward. Device. During execution of a sequence in which sequence steps transition sequentially, the transition timing of the sequence step is acquired, and the preceding transition timing is specified as the preceding transition timing from among the already acquired transition timings, and the preceding transition timing is determined from the preceding transition timing. , an acquisition step of acquiring the elapsed time until the operation of the operated device as the operation timing time;
A reference value calculated based on a history operation timing time, which is a past operation timing time, for each operation of a plurality of devices that operated during past execution of the sequence is stored in a reference value storage unit in association with the operation of the device. a reference value storage step for
specifying the reference value of the motion for which the operation timing time is acquired in the acquisition step from the reference value storage unit in response to acquisition of the operation timing time, and comparing the specified reference value and the operation timing time; a dissimilarity calculation step of calculating the dissimilarity of
and an image generating step of generating an image representing the degree of difference and displaying it on a display unit.

Images