JP6963881B2 - Voice recognition monitoring device - Google Patents

Description

An embodiment of the present invention relates to a voice recognition monitoring device that monitors each process line of a plant and enables the line to be stopped by voice.

Conventionally, the process lines of plants such as steel, non-ferrous metals, papermaking, petrochemicals, and pharmaceuticals have been monitored by HMI-based monitoring equipment. When an abnormality occurs in the process line and the process line is stopped urgently, the operator of the monitoring device needs to operate the operation supplies such as the JOG operation switch and the stop switch installed on the operation panel.

If the operator is away from the operation panel for some reason and the process line needs to be stopped urgently, there will be a time lag before returning to the operation panel and operating the operation equipment to stop the line. It will occur. In addition, since the operator is often located away from the operation panel on the machine side at the site, the stop operation may be delayed by the movement time to the operation panel, and there may be a case where the operator cannot stop even if he / she wants to stop urgently. ..

Therefore, it is desired to stop the process line by using the voice recognition technology even when the operator is at a position away from the operation panel.

On the other hand, if the operator is not proficient and is unfamiliar with judgment and operation, if the process line is accidentally stopped, it will take man-hours such as long-term adjustment to restart the process line, and the line will be down. A large amount of money may be incurred during the period.

JP-A-2002-73125 Japanese Unexamined Patent Publication No. 2002-267252

The embodiment provides a voice recognition monitoring device that quickly and reliably makes an emergency stop on a process line while avoiding an accidental stop.

The voice recognition monitoring device according to the embodiment voice-stops the operation of at least a part of the process line in the plant. This voice recognition monitoring device includes a first sound collecting means for inputting voice and generating voice data, a storage device for storing preset speaker voice data as registration data, and the voice data being registered. It is determined whether or not it corresponds to the data, and if the voice data corresponds to the registered data, the voice data is processed, and if the voice data does not correspond to the registered data, the voice data is processed. A determination device for stopping the processing of the audio data is provided. The determination device determines whether or not the voice data includes a word related to stop when processing the voice data, and when the voice data includes a word related to stop, a stop command to a process control device that controls the operation of the plant. Generates a control signal containing. The first sound collecting means includes a switch for setting operation and cutoff. The storage device includes the registration data associated with each of a plurality of operator categories set according to the processing that can be executed by the determination device. When the voice data corresponds to one of the plurality of operator categories, the determination device continues processing of the voice data regardless of the operation and stop of the switch, and the voice data is the plurality of operator categories. When corresponding to the other one, the processing of the voice data is continued when the operation of the switch is selected, and the processing of the voice data is stopped when the shutoff of the switch is selected.

In the present embodiment, a voice recognition monitoring device that quickly and surely makes an emergency stop of a process line while avoiding an erroneous stop is realized.

FIG. 1A is a schematic plan view illustrating the voice recognition monitoring device according to the embodiment. FIG. 1B is a schematic front view illustrating the voice recognition monitoring device according to the embodiment. This is an example of a flowchart for explaining the operation of the voice recognition monitoring device of the embodiment. 3 (a) and 3 (b) are schematic perspective views illustrating a part of the voice recognition monitoring device of the modified example of the embodiment.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
The drawings are schematic or conceptual, and the relationship between the thickness and width of each part, the ratio of the sizes between the parts, and the like are not necessarily the same as the actual ones. Further, even when the same parts are represented, the dimensions and ratios may be different from each other depending on the drawings.
In the specification of the present application and each of the drawings, the same elements as those described above with respect to the above-described drawings are designated by the same reference numerals, and detailed description thereof will be omitted as appropriate.

As shown in FIG. 1A, the voice recognition monitoring device 4 of the embodiment is installed in the operation room 1. An operation panel 2 is installed in the operation panel 1, and the voice recognition monitoring device 4 is arranged adjacent to the operation panel 2 and is used together with the operation panel 2.

The operation room 1 is provided with a door 1a, and an operator or other person who is permitted to enter the operation room 1, for example, can enter and exit the operation room 1. In this example, the operation room 1 is provided with a monitoring screen (not shown) for monitoring the state of various lines of the plant on the wall surface 1b facing the door 1a. Further, the situation of the site outside the operation room 1 may be visually recognized from the window provided on the wall surface 1b. The operator can recognize the status of various process lines of the plant by monitoring screen or visual inspection.

An operation panel switch 9, an operation panel lamp 6, an operation panel JOG switch 7, and the like are arranged on the operation panel 2. Although not shown, the operation panel 2 is connected to a process controller (process control device) via, for example, a high-speed network for control. The process controller acquires, for example, the operating status of devices and devices arranged on the process line, various data related to the surrounding environment, and the like based on a program. The process controller controls the process line by generating and transmitting various commands and control signals to these devices and devices based on the acquired various data and programs. The operator can intervene in the control of the process controller by operating the operating equipment such as the switch on the operation panel 2. The operator operates these switches and the like on the operation panel 2 according to the monitoring screen and the visual situation so that the process line can be smoothly controlled.

The voice recognition monitoring device 4 is connected to the process controller via a control network. The voice recognition monitoring device 4 can acquire the voice of the operator and intervene in the control of the process line without going through the operation by the operation panel 2. Here, the control of the process line by the voice recognition monitoring device 4 is the stop of the process line. An operation button or the like for emergency stop of the process line is also provided on the operation panel 2, and the process line can be stopped manually by the operator.

The voice recognition monitoring device 4 includes a microphone (first sound collecting means) 3, and the microphone 3 is provided on the voice recognition monitoring device 4 together with the microphone switch 8. The microphone 3 and the microphone switch 8 are electrically connected to the voice recognition monitoring device 4. In this example, the microphone 3 and the microphone switch 8 are fixed on the top plate of the voice recognition monitoring device 4. In this example, the microphone 3 and the microphone switch 8 are separately provided on the top plate of the voice recognition monitoring device 4, but the microphone switch 8 may be provided on the microphone 3. Further, as described above, it is assumed that the microphone 3 and the microphone switch 8 are electrically connected via the voice recognition monitoring device 4, and the microphone switch 8 is substantially included in the microphone 3. ..

The microphone 3 is, for example, omnidirectional, and can acquire the voice emitted by the operator and supply the acquired voice data to the voice recognition monitoring device 4 regardless of where the operator is in the operation room 1.

The microphone switch 8 is a push button type switch, and is turned on when the microphone switch 8 is pressed, and turned off when the microphone switch 8 is stopped pressed.

The voice recognition monitoring device 4 can input the voice data acquired by the microphone 3 by turning on the microphone switch 8. The voice recognition monitoring device 4 can stop the input of the acquired voice data by turning off the microphone switch 8. As will be described in detail later, the voice data in the voice recognition monitoring device 4 is preset according to the operator classification as to how it is handled by turning on / off the microphone switch 8.

The voice recognition monitoring device 4 recognizes the input voice data and generates a control signal based on the content of the voice data. The control signal is a signal including a process line stop command. The voice data can include data representing a section of the process line, and the voice recognition monitoring device 4 generates a control signal including a stop command in the recognized section. Since the voice recognition monitoring device 4 is connected to the process controller, the generated control signal is transmitted to the process controller, and the process controller operates to stop the corresponding section of the process line.

A process line section represents an area or location in one process line. For example, in a steel sheet rolling line, in a transport area before the rolled material is put into the rolling mill, an area where the rolled material is rolled by the rolling mill, a transport area after the rolled material is discharged from the rolling mill, and the like. Yes, it is preset according to the process line.

The voice recognition monitoring device 4 includes a determination device 4a and a storage device 4b. The determination device 4a and the storage device 4b are connected to each other. The determination device 4a may include an arithmetic unit that operates according to a program such as a CPU (Central Processing Unit) of a computer device. A program for operating the determination device 4a is stored in the storage device 4b connected to the determination device 4a. The determination device 4a operates according to each step of the program read from the storage device 4b. The determination device 4a generates a control signal including a stop command based on the recognized voice and transmits it to the process controller.

The determination device 4a identifies the speaker based on the voice data acquired by the microphone 3. The speaker is registered in advance, and the registered data is stored in the storage device 4b. When the speaker of the acquired voice data matches the speaker of the registered data, the determination device 4a can continue the process based on the acquired voice data. If the speaker of the acquired voice data does not match the speaker of the registered data, the determination device 4a stops processing the voice data regardless of whether the microphone switch 8 is turned on or off.

A plurality of operator categories are set in the registration data related to the speaker. The operator classification is set according to the proficiency level of the operation of the device or the like in the operation room 1 including the operation panel 2 of the operator, for example. The determination device 4a executes the process according to the operator classification.

The processing based on the operator classification is set in two stages according to, for example, the proficiency level of the operator. The operator category "1" applies to the most proficient operator. When the operator classification is "1", for example, regardless of whether the microphone switch 8 is turned on or off, the voice recognition monitoring device 4 executes the processing of the operator's voice data and outputs a control signal including a stop command. Generate. The operator category "2" applies to unskilled operators. When the operator classification is "2", for example, the voice recognition monitoring device 4 processes the voice data only when the microphone switch 8 is on. When the microphone switch 8 is off, the voice recognition monitoring device 4 stops processing the voice data.

The voice data can include a plurality of types of words. One of the multiple types of words is the word for "stop" and the other is the word for "section" of the process line. In order to continue processing the audio data, it may be mandatory for the audio data to include a word for "stop", and it may be optional to include a word for "section". In the following, a case where the voice data needs to include a word related to "stop" in order to continue processing the voice data will be described.

The determination device 4a analyzes the voice data, and if the word related to "stop" is included in the voice data, generates a control signal including a stop command and transmits the control signal to the process controller. As for the word related to "stop", for example, in addition to "stop", "stop", "stop" and the like are registered in the storage device 4b in advance. If the voice data does not include these "stop" words, the determination device 4a does not generate a control signal including a stop command regardless of the presence or absence of other words.

The determination device 4a analyzes the voice data of the speaker, and if the word related to the "section" is included in the data, associates the section data with the stop command to generate a control signal, and the process controller. Send to. Words relating to "section" are, for example, "inside", "input", "center", "center", "outside", "output" and the like. In addition to these, the word related to the "section" is registered in the storage device 4b in advance according to the line. Also, on the process controller side, "sections" are assigned in advance to devices and devices.

The operation of the voice recognition monitoring device 4 of the embodiment will be described.
FIG. 2 is an example of a flowchart illustrating the operation of the voice recognition control device of the embodiment.
As in the above example, it is assumed that the operator categories "1" (highly proficient) and "2" (lowly proficient) are set.

As shown in FIG. 2, when a situation to be stopped at the process line occurs in step S0, in step S1, the operator determines whether or not the process line needs to be stopped. If it is determined that the stop is necessary, the operator speaks into the microphone 3 in step S2. In this case, the operator corresponding to the operator category "2" turns on the microphone switch 8 and speaks. In the case of an operator of the operator category "1" or an outsider who is permitted to enter the room, the microphone switch 8 may be turned on or may be left off.

In step S3, the determination device 4a analyzes the voice data acquired by the microphone 3 and identifies the speaker with reference to the registered data. If the speaker matches the speaker of the registered data, the operator classification of the registered data is set. If the speaker does not match the speaker of the registered data, the determination device 4a does not execute the subsequent processing and waits until other voice data is input.

In step S4, the determination device 4a sets the process according to the operator classification of the voice data. In the case of the operator category "1", the determination device 4a continues the analysis of the voice data regardless of whether the microphone switch 8 is on or off, and determines the presence or absence of the registered word. In the case of the operator category “2”, the determination device 4a continues the analysis of the voice data when the microphone switch 8 is on. When the microphone switch 8 is off, the determination device 4a stops the analysis of the voice data until the microphone switch 8 is turned on.

In step S5, the determination device 4a analyzes the voice data and determines whether or not the voice data includes a word related to "stop". Further, the determination device 4a determines whether or not the voice data includes a word related to the "section". If the voice data does not include a word related to the "section", the determination device 4a shifts the process to step S6. When the voice data includes data related to the "section", the determination device 4a shifts the process to step S7. If the voice data does not include a word related to "stop", it waits until other voice data is acquired.

In step S6, the determination device 4a generates a control signal to stop the operation of all sections of the process line when the voice data includes a word related to "stop" and does not include a word related to "section". ..

In step S7, when the voice data includes both a word related to "stop" and a word related to "section", the determination device 4a generates a control signal including a stop command for the "section".

In step S8, if the voice data does not include a word related to "stop", the determination device 4a does not generate a control signal and waits until another voice data is input and processed.

The voice data may include a word related to "stop" and a word related to "section", and the word order may be set regardless of the word order. For example, the word "stop" may be enabled and vice versa only if the word "stop" precedes the word "section". In addition, if there is a silent period of a predetermined time or more between the word related to "stop" and the word related to "section", only the first word may be treated as a valid word. .. If there is a word that is neither "stop" nor "section" between the word "stop" and the word "section", exclude that word and "stop" and "section". You may want to process words related to "sections".

In the above, when the voice data does not include the word related to the "section", all the sections of the process line are stopped, but it can be set appropriately according to the type of the process line and the like. For example, when the voice data does not include the word related to the "section", the process may be stopped and wait until the voice data including the word related to the "section" is input.

Further, in the above description, when the voice data does not include the word related to "stop", the determination device 4a is supposed to stop the processing of the voice data. It can be set appropriately according to the situation. For example, the determination device 4a may stop processing the voice data when the voice data does not contain either a word related to "stop" and a word related to "section". In that case, when the voice data contains only the word related to the "section" and does not include the word related to the "stop", the determination device 4a is made to generate a control signal including the stop command for the "section". You may.

(Modification example)
3 (a) and 3 (b) are schematic perspective views illustrating a part of the voice recognition control device of the modified example of the embodiment.
FIG. 3A shows an outline of a portable wired switch for a microphone. In the above-described embodiment, the microphone switch 8 is fixed to the voice recognition monitoring device 4, but in this modified example, the operator carries the microphone switch 18 in his / her hand and moves in the operation room 1. can.

As shown in FIG. 3A, the portable microphone switch 18 includes a main body 18a and a switch 18b. The microphone switch 18 is electrically connected to the voice recognition monitoring device 4 by the wiring 18c. In this example, the microphone switch is connected to the voice recognition monitoring device 4 by wire, but of course, it may be connected wirelessly. The main body 18a has a cylindrical shape with a bottom, and a push button type circular switch 18b is provided on the ceiling portion. The thickness of the main body 18a is such that when an adult holds it with one hand, there is room in the palm.

When the operator enters the operation room 1 and works, he / she usually stands or sits in front of the operation panel 2 (FIG. 1) to perform various operations, but when he / she leaves the operation panel 2, he / she performs various operations. Hold the microphone switch 18 in your hand and carry it.

The operation of the microphone switch 18 is the same as that of the microphone switch 8 in the above-described embodiment. That is, an operator having an operator classification in which processing of voice data is permitted by turning on the microphone switches 8 and 18 turns on the microphone switches 8 and 18 to perform a voice stop operation when an emergency stop is required. can do.

By making the microphone switch 18 portable, it is possible to perform a stop operation by voice on the spot without moving to the microphone switch 8 fixed on the voice recognition monitoring device 4. Therefore, even an operator with a low level of proficiency can quickly perform a stop operation.

FIG. 3B shows an outline of the portable microphone 23. The operator may leave the operation room 1 and perform inspection, monitoring work, etc. of the equipment, the equipment, etc. in the vicinity of the equipment, the equipment, etc. at the site. An operation panel for equipment and devices is provided at the site, and the equipment and devices can be operated. However, in this modified example, voice is provided without going through the operation panel on the machine side in an emergency or emergency. Can stop the process line.

As shown in FIG. 3B, the microphone (second sound collecting means) 23 includes a wireless connection device 24 and a microphone switch 28. The wireless connection device 24 transmits the voice data acquired by the microphone 23 to the voice recognition monitoring device 4. Similar to the microphone switch 8 of the embodiment, the microphone switch 28 is a microphone switch when an operator having an operator classification in which processing of voice data is permitted by turning on the microphone switch 28 needs an emergency stop. Turn on 28 to perform a voice stop operation.

The effect of the voice recognition monitoring device 4 of the embodiment and the modified example will be described.
In the voice recognition monitoring device 4 of the embodiment and the modified example, it is determined whether or not the determination device 4a is a speaker registered in the storage device 4b in advance. Therefore, the process line is not accidentally stopped by the voice in the operation room 1 in which a person other than the operator such as a visitor enters and exits.

In the voice recognition monitoring device 4, since the process line can be stopped by voice, even an operator who is not proficient in the operation of the operation panel 2 can stop the process line reliably and quickly.

In the voice recognition monitoring device 4, a plurality of categories can be set for the registration data to be registered as a speaker. A plurality of divisions can be set according to the proficiency level of the operator, and multiple protective measures for the stop operation can be taken for the operator with a low proficiency level. As a protective measure, microphone switches 8, 18 and 28 are provided to prevent an unskilled operator from accidentally stopping the process line.

For example, in a steel plant, if an abnormality occurs in the rolling process, if emergency stop is not taken and appropriate measures are not taken, the damage will spread, it will take a long time to restart, and a large cost will be required. Become. In the voice recognition monitoring device 4 of the embodiment and the modified example, the process line can be stopped by voice, so that the process line can be stopped quickly without waiting for the operation of the operation panel 2. Even an operator with a low level of proficiency who is unfamiliar with the operation procedure of the operation panel 2 can reliably perform processing at the time of abnormality.

On the other hand, in the voice recognition monitoring device 4 of the embodiment and the modified example, since it is possible to determine the proficiency level of the operator and the presence / absence of operation permission from the voice data, an inappropriate stop is mistakenly made, and the stop is not originally necessary. It is possible to prevent the loss from occurring.

According to the embodiment described above, it is possible to realize a voice recognition monitoring device that makes an emergency stop quickly and surely while avoiding an erroneous stop of the process line.

Although some embodiments of the present invention have been described above, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other embodiments, and various omissions, replacements, and changes can be made without departing from the gist of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are also included in the scope of the invention and its equivalents described in the claims. Moreover, each of the above-described embodiments can be implemented in combination with each other.

1 operation room, 2 operation panel, 3,23 microphones, 4 voice recognition monitoring device, 4a judgment device, 4b storage device, 8,18,28 microphone switch, 24 wireless connection device

Claims (5)

A voice recognition monitoring device that stops the operation of at least a part of the process line in the plant by voice.
The first sound collecting means for inputting voice and generating voice data,
A storage device that stores preset speaker voice data as registered data,
It is determined whether or not the voice data corresponds to the registered data, and if the voice data corresponds to the registered data, the voice data is processed, and the voice data corresponds to the registered data. If it is not the case, the judgment device that stops the processing of the voice data and
With
The determination device determines whether or not the voice data includes a word related to stop when processing the voice data, and when the voice data includes a word related to stop, a stop command to a process control device that controls the operation of the plant. It generates a control signal including,
The first sound collecting means includes a switch for setting operation and cutoff.
The storage device includes the registration data associated with each of the plurality of operator categories set according to the processing that can be executed by the determination device.
When the voice data corresponds to one of the plurality of operator categories, the determination device continues processing of the voice data regardless of the operation and stop of the switch.
When the voice data corresponds to the other one of the plurality of operator categories, the processing of the voice data is continued when the operation of the switch is selected, and the voice is selected when the shutoff of the switch is selected. A voice recognition monitoring device that stops processing data. The process line includes a plurality of sections representing areas or locations of the process line.
The judgment device is
If the processing of the voice data is continued,
Determine if the audio data contains words for at least one of the plurality of sections.
When the voice data contains words relating to at least one of the plurality of sections, a control signal including a stop command for that section is generated.
Wherein when the speech data does not include words relating to any of said plurality of sections, the plurality of claim 1 Symbol placement of the speech recognition monitoring device generates a control signal including a stop instruction to all sections. The process line includes a plurality of sections representing areas or locations of the process line.
The judgment device is
If the processing of the voice data is continued,
Determine if the audio data contains words for at least one of the plurality of sections.
When the voice data contains words relating to at least one of the plurality of sections, a control signal including a stop command for that section is generated.
Wherein when the speech data does not include words relating to any of said plurality of sections, the process control according to claim 1 Symbol placement of the speech recognition monitoring device to continue the operation of the device. The switch is portable and is claim 1 speech recognition monitoring device as claimed. The voice recognition monitoring device according to any one of claims 1 to 4 , further comprising a second sound collecting means that can be wirelessly connected to the determination device.

Images