JP7056753B2 - Elevator brake device abnormality diagnosis system - Google Patents

Description

The present invention relates to an elevator brake device abnormality diagnosis system.

Patent Document 1 describes an example of an elevator brake device abnormality diagnosis system. The brake device abnormality diagnosis system measures the stroke of the plunger of the brake device with a laser displacement meter. The brake device abnormality diagnosis system diagnoses the brake device as an abnormality when the measured stroke reaches a threshold value.

Japanese Patent Application Laid-Open No. 2015-42892

However, the brake device abnormality diagnosis system of Patent Document 1 determines the abnormality of the brake device based on a predetermined constant threshold value with respect to the stroke of the plunger. Therefore, the brake device abnormality diagnosis system cannot diagnose the abnormality of the brake device when the threshold value for determining the abnormality is unknown.

The present invention has been made to solve such a problem. An object of the present invention is to provide an abnormality diagnosis system capable of diagnosing an abnormality of a braking device based on data having an unknown threshold value for diagnosing an abnormality.

The brake device abnormality diagnosis system for an elevator according to the present invention has an observation unit that acquires operation data about the operation of the brake device when the brake device that brakes the car of the elevator operates, and an operation data acquired by the observation unit. A conversion unit that converts to state data corresponding to the failure phenomenon of the brake device, and a method of unsupervised learning using state data, or a method of supervised learning using judgment data and state data that determine abnormalities in the brake device. The brake device is based on the diagnostic model from the state data obtained by the conversion unit converting the operation data acquired by the observation unit after learning by the learning unit and the learning unit that learns the diagnostic model of the abnormality of the brake device. A determination unit for determining an abnormality is provided , and the determination unit is at the same position as the position of the elevator car when the determination is made when the result of the abnormality determination of the brake device represents an abnormal state. A signal for braking the car is output to the braking device, and the state data obtained by converting the operation data acquired by the observation unit when the braking device brakes the car at the position is converted by the conversion unit to brake based on the diagnostic model. Judge an abnormality in the device .
The brake device abnormality diagnosis system for an elevator according to the present invention has an observation unit that acquires operation data about the operation of the brake device when the brake device that brakes the car of the elevator operates, and an operation data acquired by the observation unit. A conversion unit that converts to state data corresponding to the failure phenomenon of the brake device, and a method of unsupervised learning using state data, or a method of supervised learning using judgment data and state data that determine abnormalities in the brake device. Brake device abnormality based on the diagnostic model from the learning unit that learns the diagnostic model of the brake device abnormality by The determination unit is provided with a determination unit for determining an abnormality, and the determination unit brakes at a position different from the position of the elevator car when the determination is made when the result of the abnormality determination of the braking device indicates an abnormal state. A braking device is output based on a diagnostic model from the state data obtained by outputting a signal to brake the car to the device and converting the operation data acquired by the observation unit when the braking device brakes the car at the position. Judge the abnormality of.
The elevator brake device abnormality diagnosis system according to the present invention has an observation unit that acquires operation data about the operation of the brake device when the brake device that brakes the elevator car operates, and an operation data acquired by the observation unit. A conversion unit that converts to state data corresponding to the failure phenomenon of the brake device, a method of unsupervised learning using state data, or a method of supervised learning using judgment data and state data that determine abnormalities in the brake device. Based on the diagnostic model, the learning unit learns the diagnostic model of the abnormality of the brake device by, and the conversion unit converts the operation data acquired by the observation unit after the learning by the learning unit. A determination unit that determines The unit classifies the operation data based on the environmental data and the deterioration time predicted by the prediction unit, the learning unit learns the diagnostic model for each classification by the classification unit, and the judgment unit learns for each classification by the classification unit. The abnormality of the braking device is judged based on the diagnostic model made.

According to these inventions, the brake device abnormality diagnosis system includes an observation unit, a conversion unit, a learning unit, and a determination unit. When the brake device that brakes the car of the elevator operates, the observation unit acquires operation data about the operation of the brake device. The conversion unit converts the operation data acquired by the observation unit into state data corresponding to the failure phenomenon of the brake device. The learning unit learns a diagnostic model of the abnormality of the brake device by the method of supervised learning or unsupervised learning using the state data. The determination unit determines the abnormality of the brake device based on the diagnostic model from the state data obtained by converting the operation data acquired by the observation unit after the learning by the learning unit by the conversion unit. As a result, it is possible to diagnose the abnormality of the brake device based on the data whose threshold value for diagnosing the abnormality is unknown.

It is a block diagram of the brake device abnormality diagnosis system which concerns on Embodiment 1. FIG. It is a figure which shows the example of the abnormality diagnosis by the brake device abnormality diagnosis system which concerns on Embodiment 1. FIG. It is a flowchart which shows the example of the operation of the brake device abnormality diagnosis system which concerns on Embodiment 1. FIG. It is a flowchart which shows the example of the operation of the brake device abnormality diagnosis system which concerns on Embodiment 1. FIG. It is a figure which shows the example of the abnormality diagnosis by the brake device abnormality diagnosis system which concerns on Embodiment 1. FIG. It is a figure which shows the hardware composition of the main part of the brake device abnormality diagnosis system which concerns on Embodiment 1. FIG.

The embodiment for carrying out the present invention will be described with reference to the accompanying drawings. In each figure, the same or corresponding parts are designated by the same reference numerals, and duplicate description will be appropriately simplified or omitted.

Embodiment 1.
FIG. 1 is a configuration diagram of a brake device abnormality diagnosis system 1 according to the first embodiment.

The brake device abnormality diagnosis system 1 is applied to the elevator 2.

The elevator 2 is provided in the building 3. Building 3 has a plurality of floors. In the elevator 2, the hoistway 4 penetrates each floor of the building 3. In the elevator 2, the landing 5 is provided on each floor of the building 3. The landing 5 on each floor faces the hoistway 4. In the elevator 2, each of the plurality of landing doors 6 is provided at the landing 5 on each floor. The elevator 2 includes a hoisting machine 7, a main rope 8, a balance weight 9, a car 10, a brake device 11, a control panel 12, and a monitoring device 13.

The hoisting machine 7 is provided, for example, on the upper part of the hoistway 4. The hoist 7 includes a motor and a sheave. The motor of the hoist 7 is a device for rotating the sheave.

The main rope 8 is wound around the sheave of the hoist 7 so that it can move following the rotation of the sheave of the hoist 7. One end of the main rope 8 is provided in the car 10. The other end of the main rope 8 is provided on the balanced weight 9.

The balance weight 9 is provided so as to be able to travel in the vertical direction inside the hoistway 4 following the movement of the main rope 8.

The car 10 is provided so as to be able to travel in the vertical direction inside the hoistway 4 following the movement of the main rope 8. The car 10 includes a car door 14. The car door 14 is a device that opens and closes when the car 10 is stopped on any of the floors of the building 3. The car door 14 is a device that opens and closes the landing door 6 in conjunction with each other.

The brake device 11 is a device that brakes the car 10 when the car 10 is stopped. The brake device 11 includes a brake drum 15, a brake shoe 16, a coil 17, a plunger 18, a spring 19, and a brake control device 20. The brake drum 15 is provided on the output shaft of the motor of the hoisting machine 7 so that it can rotate in synchronization with the motor of the hoisting machine 7. The brake shoe 16 faces the outer surface of the brake drum 15. The brake shoe 16 is a device that brakes the car 10 by braking the rotation of the brake drum 15 by a frictional force. The spring 19 is a device that presses the brake shoe 16 against the brake drum 15 by an elastic force. The coil 17 is a device that generates a magnetic field by energization. The plunger 18 is a device that displaces the brake shoe 16 away from the brake drum 15 while resisting the elastic force of the spring 19 by the magnetic field generated by the coil 17. The brake control device 20 is a device that controls the operation of the brake device 11. The operation of the braking device 11 includes suction and release. The brake control device 20 is equipped with an element that outputs a suction command and a release command. The suction command is output when the braking device 11 brakes the car 10. The release command is output when the braking device 11 brakes the car 10. The brake device 11 may include a brake arm that transmits the elastic force of the spring 19 to the brake shoe 16.

The control panel 12 is provided, for example, on the upper part of the hoistway 4. The control panel 12 is a device that controls the operation of the elevator 2. The operation of the elevator 2 includes, for example, traveling of the car 10. The control panel 12 is connected to the hoisting machine 7 and the brake device 11 so as to be able to control the operation of the elevator 2.

The monitoring device 13 is provided in, for example, a building 3. The monitoring device 13 is a device that monitors the operation of the elevator 2. The monitoring device 13 is connected to the control panel 12 so as to be able to communicate data about the operation of the elevator 2.

The elevator 2 is provided with an operation measuring device (not shown) and an environmental measuring device.

The motion measuring device is a device that acquires motion measurement data when the brake device 11 operates. The motion measurement data is multi-component data representing information about the motion of the brake device 11. A part or all of the motion measuring device is provided in, for example, a brake device 11, a hoisting machine 7, or a car 10. The motion measuring device includes, for example, a sensor, a switch, and the like. The motion measuring device includes, for example, an ammeter, a brake switch, and an encoder.

The ammeter is provided, for example, in the wiring that supplies electric power to the coil 17. The ammeter is a sensor that measures the current applied to the coil 17. The brake switch is provided on the brake device 11. The brake switch is a switch that detects the operating state of the brake device 11. The operating state of the brake device 11 includes a braking state and a released state. The brake switch includes, for example, a mechanism for detecting the operating state of the brake device 11 by detecting the mechanical displacement of a part of the brake device 11. The encoder is provided in the motor of the hoist 7. The encoder is a sensor that outputs the rotation angle of the motor of the hoist 7 by a pulse signal.

Information on each component of the motion measurement data is output to the control panel 12. Alternatively, the information of each component of the motion measurement data is output to the control panel 12 through the brake control device 20. The control panel 12 stores the motion measurement data together with the signal data and the calculation data so that it can be output as motion data. The signal data is multi-component data representing information on the presence / absence of input or output of a control signal. The control signals are, for example, a brake voltage command, a suction command, a release command, a brake voltage command and a brake contact signal. The variables of the control software may include information of calculated data. The calculated data is multi-component data calculated based on motion measurement data, signal data, and the like.

The environmental measurement device is a device that acquires environmental measurement data. The environment measurement data is multi-component data representing information about the operating environment of the brake device 11. A part or all of the environment measuring device is provided in, for example, a brake device 11, a hoisting machine 7, or a car 10. The environment measuring device is provided, for example, in the hoistway 4. The plurality of environmental measuring devices include, for example, a scale and a thermometer.

The scale is provided in the basket 10. The scale is a sensor that measures the weight of a user or the like in the car 10. The thermometer is provided in the hoistway 4. A thermometer is, for example, a sensor that measures air temperature. The thermometer may be provided in the brake device 11. At this time, the thermometer is, for example, a sensor that measures the temperature of the brake device 11.

Information on each component of the environmental measurement data is output to the control panel 12. Alternatively, the information of each component of the environmental measurement data is output to the control panel 12 through the brake control device 20. The control panel 12 stores the environment measurement data so that it can be output.

In the brake device abnormality diagnosis system 1, the information center 21 is provided, for example, outside the building 3. The information center 21 is a base for collecting information on the elevator 2 and other elevators.

The brake device abnormality diagnosis system 1 is a system for diagnosing an abnormality of the brake device 11. The brake device abnormality diagnosis system 1 may have a function of predicting the deterioration time of the brake device 11.

The brake device abnormality diagnosis system 1 includes a data server 22, a maintenance support device 23, and a display device 24.

The data server 22 is provided, for example, in the information center 21. The data server 22 is connected to the monitoring device 13 so as to be able to communicate information such as the operation of the elevator 2. The data server 22 includes an observation data storage unit 25, an attribute data storage unit 26, and an abnormal data storage unit 27.

The observation data storage unit 25 is a portion that stores the observation database. The observation database contains multiple observation data. Observation data includes operation data and environmental measurement data.

The attribute data storage unit 26 is a portion that stores the attribute database. The attribute database contains multiple attribute data. The attribute data includes data based on the attributes of the elevator. Further, the attribute data includes data based on the attributes of the braking device. The attribute data includes information such as, for example, the model of the brake device, the device weight of the car, the type of the elevator, and the area where the elevator is installed. The type of elevator includes information such as whether or not it is a viewing elevator. The type of elevator relates, for example, to the environment of the hoistway. The type of elevator relates, for example, to the model of the elevator. The area where the elevator is installed is related to the environment of the hoistway, for example through the climate. The area where the elevator is installed is related to the environment of the hoistway, for example through the concentration of salt or sulfur in the air.

The abnormality data storage unit 27 is a portion that stores the abnormality history database. The anomaly history database contains a plurality of determination data for elevator 2 and other elevators. The determination data is data for determining an abnormality in the brake device 11. The determination data includes, for example, information on the presence or absence of an abnormality, the type of abnormality, and the degree of abnormality. The determination data is, for example, data associated with one operation data.

The maintenance support device 23 is provided in, for example, the information center 21. The maintenance support device 23 includes an observation unit 28, a data acquisition unit 29, a classification unit 30, a conversion unit 31, a learning unit 32, a determination unit 33, a generation unit 34, a prediction unit 35, and a storage unit 36. And a notification unit 37.

The observation unit 28 is a part that acquires operation data when the brake device 11 operates. The observation unit 28 is connected to the monitoring device 13 so that observation data including operation data can be acquired.

The data acquisition unit 29 is a part that generates a training data set. The training data set contains multiple sets of environmental data, operation data and judgment data. Environmental data includes environmental measurement data and attribute data. The data acquisition unit 29 is connected to the observation data storage unit 25 so that observation data can be acquired. The data acquisition unit 29 is connected to the attribute data storage unit 26 so that the attribute data can be acquired. The data acquisition unit 29 is connected to the abnormal data storage unit 27 so that determination data can be acquired.

The classification unit 30 is a part that classifies operation data based on environmental data. The classification unit 30 is connected to the observation unit 28 so that operation data can be acquired. The classification unit 30 is connected to the observation unit 28 and the attribute data storage unit 26 so that the environment measurement data and the attribute data can be acquired as the environment data. The classification unit 30 is connected to the data acquisition unit 29 so that the training data set can be acquired.

The conversion unit 31 is a part that converts operation data into state data and index data.

The state data is multi-component data. Each component of the state data corresponds to each failure phenomenon of the brake device 11. Each failure phenomenon of the brake device 11 includes, for example, sticking of the contact of the relay switch, deterioration of the spring 19, misalignment of the brake shoe 16, deterioration of the braking ability of the brake device 11, and abnormality of the electronic circuit of the brake control device 20. including.

The index data is data representing deterioration of the brake device 11. The index data is, for example, time-series data representing a deterioration index value for each time unit set in advance. The deterioration index value is a value that is an index indicating the deterioration of the brake device 11. The deterioration index value may be a multi-component value. The deterioration of the brake device 11 is, for example, the wear of the brake shoe 16. Deterioration of the brake device 11 reduces, for example, the braking ability of the brake device 11. A decrease in the braking ability of the brake device 11 causes, for example, slip to occur in the brake device 11. The time unit of the time series data is, for example, one day. The conversion unit 31 is connected to the classification unit 30 so that operation data classified based on the environmental data can be acquired.

The learning unit 32 is a part that learns a diagnostic model of an abnormality of the brake device 11 using the state data. The learning method by the learning unit 32 is a machine learning method. The learning unit 32 is connected to the conversion unit 31 so that state data can be acquired. The learning by the learning unit 32 is performed, for example, by an operation of starting learning by the operator of the information center 21.

The determination unit 33 has an abnormality in the brake device 11 based on the diagnostic model learned by the learning unit 32 from the state data obtained by converting the operation data acquired by the observation unit 28 after the learning by the learning unit 32 by the conversion unit 31. Is the part to judge. The determination unit 33 is connected to the conversion unit 31 so that state data can be acquired. The determination unit 33 is connected to the learning unit 32 so that the diagnostic model can be acquired. The determination by the determination unit 33 is performed each time the state data is acquired, for example, when the determination unit 33 is activated. The determination unit 33 is activated, for example, by an operation of activation by the operator of the information center 21. The determination unit 33 is connected to the monitoring device 13 so that the determination result can be output.

The generation unit 34 is a part that generates a deterioration model representing a change with respect to the time of deterioration represented by the index data. The deterioration model is a model that predicts future changes in the deterioration index value. The degradation model includes a trend component, a periodic component and a short-term fluctuation component. A trend component is a component that represents a long-term trend of monotonous changes in increase or decrease. The periodic component is a component that represents the tendency of periodic change. The short-term fluctuation component is a component representing short-term fluctuation. The generation unit 34 is connected to the conversion unit 31 so that the index data can be acquired.

The prediction unit 35 is a part that predicts the deterioration time of the brake device 11 based on the deterioration model generated by the generation unit 34. The deterioration time of the brake device 11 is a time when the deterioration index value reaches a preset threshold value. The prediction unit 35 is connected to the generation unit 34 so that the deterioration model can be read.

The storage unit 36 is a unit for storing determination result data. The determination result data is data representing the result of the determination by the determination unit 33. The storage unit 36 is connected to the determination unit 33 so that the determination result data can be acquired. The storage unit 36 is a unit for storing prediction result data. The prediction result data is data representing the result of prediction by the prediction unit 35. The storage unit 36 is connected to the prediction unit 35 so that the prediction result data can be acquired.

The notification unit 37 is a unit that notifies the result of the determination of the abnormality of the brake device 11 by the determination unit 33. The notification unit 37 is connected to the determination unit 33 so that the determination result data can be acquired. The notification unit 37 is a unit that notifies the result of prediction of the deterioration time of the brake device 11 by the prediction unit 35. The notification unit 37 is connected to the prediction unit 35 so that the prediction result data can be acquired. The notification unit 37 generates notification data from the determination result data or the prediction result data. The notification data is data representing the content to be notified.

The display device 24 is a device that displays the contents represented by the acquired data. The display device 24 is, for example, a display. The display device 24 is provided in, for example, the information center 21. The display device 24 is connected to the notification unit 37 so that the notification data can be acquired.

Subsequently, the function of the brake device abnormality diagnosis system 1 will be described with reference to FIG.
FIG. 2 is a diagram showing an example of abnormality diagnosis by the brake device abnormality diagnosis system according to the first embodiment.

In the graph A, an example of the data included in the operation data is shown. The horizontal axis of graph A represents time. The vertical axis of the graph A represents the signal value measured by the motion measuring device. In Graph A, each curve represents data acquired by a single operation of the braking device 11.

The operation data is acquired as follows, for example.

The control panel 12 outputs a signal for operating the brake device 11 to the brake control device 20 when the car 10 is stopped.

The brake control device 20 operates the brake device 11 according to the control signal input from the control panel 12. When the brake device 11 operates, the motion measuring device acquires motion measurement data. The motion measuring device outputs motion measurement data to the brake control device 20 or the control panel 12. When the brake device 11 operates, the environment measurement device acquires the environment measurement data. The environment measurement device outputs the environment measurement data to the brake control device 20 or the control panel 12. The brake control device 20 outputs the input motion measurement data and environment measurement data to the control panel 12.

The control panel 12 calculates the calculated data based on the motion measurement data, the signal data, and the like. The calculated data includes, for example, the data of the position of the car 10 calculated from the count of the pulse signal of the encoder. The calculated data includes, for example, time difference data from the output of the brake suction command signal to the detection of the actual operation of the brake device 11 by the brake switch. The calculated data includes, for example, data of the time during which the braking device 11 continues the braking operation. The calculated data includes, for example, data on the frequency of operation of the brake device 11. The control panel 12 outputs motion measurement data, signal data, and calculation data as motion data to the observation unit 28 through the monitoring device 13. The control panel 12 outputs the environmental measurement data to the observation unit 28 through the monitoring device 13.

The observation unit 28 acquires operation data and environment measurement data from the control panel 12 through the monitoring device 13. The observation unit 28 outputs the operation data and the environment measurement data to the observation data storage unit 25 as observation data. The observation unit 28 outputs the operation data and the environment measurement data to the classification unit 30.

The observation data storage unit 25 stores the acquired observation data in the observation database. The observation data includes, for example, flag data, numerical data, and waveform data. The components of the operation data are, for example, flag data, numerical data, or waveform data.

The flag data includes information such as whether or not the switch has been operated, whether or not the sensor has been operated, and whether or not there is a control signal. Flag data is represented by a boolean value, an integer value, a character string, or the like.

Numerical data includes information such as, for example, the value of a physical quantity measured by a sensor. Numerical data includes, for example, the current applied to the coil 17, the duration of braking by the brake device 11, the position of the car 10, the temperature, the temperature of the brake device 11, the frequency of operation of the brake device 11, the temperature, and the car 10. Includes the weight of the user on board. Numerical data is represented by an integer value, a real value, or the like.

The waveform data includes information such as a time change of a physical quantity measured by a sensor. The waveform data includes, for example, a pattern change of the current applied to the coil 17, a time change of the position of the car 10, and a time change of the brake temperature. The waveform data is represented by a list or the like including a plurality of numerical values for each predetermined time interval.

In Graph A, an example of waveform data is shown. The plurality of curves shown in the graph A correspond to the waveform data as a component of the operation data acquired by one operation of the brake device 11, respectively.

The brake device abnormality diagnosis system 1 starts learning by, for example, an operation by an operator of the information center 21.

When the brake device abnormality diagnosis system 1 starts learning, the data acquisition unit 29 generates a training data set. The data acquisition unit 29 acquires a plurality of observation data from the observation data storage unit 25. The data acquisition unit 29 acquires a plurality of attribute data from the attribute data storage unit 26. The data acquisition unit 29 acquires a plurality of determination data from the abnormal data storage unit 27. The data acquisition unit 29 generates a plurality of operation data and a plurality of environmental data from a plurality of observation data and a plurality of attribute data. The data acquisition unit 29 associates a plurality of determination data with a plurality of operation data and a plurality of environment data. At this time, the data acquisition unit 29 may make an association using, for example, the operation time of the brake device 11. Alternatively, when the identification information is attached to one operation of the brake device 11, the data acquisition unit 29 may make an association using the identification information. The data acquisition unit 29 generates a training data set based on the correspondence. The data acquisition unit 29 outputs the training data set to the classification unit 30.

The classification unit 30 classifies the operation data corresponding to the environmental data based on the environmental data included in the training data set. For example, when the environmental data includes the labeling data for classification, the classification unit 30 classifies the operation data corresponding to the environmental data having the same value of the labeling data into the same cluster. Alternatively, the classification unit 30 classifies the operation data corresponding to the environmental data classified into the same cluster by, for example, the method of unsupervised learning into the same cluster. At this time, the classification unit 30 uses, for example, the k-means method, which is a non-hierarchical classification method, as a method of unsupervised learning. Alternatively, the classification unit 30 may use a hierarchical classification method. The classification unit 30 outputs the classified operation data to the conversion unit 31.

The conversion unit 31 converts the operation data into state data through each of the feature amount extraction step, the standardization step, the abnormality degree calculation step, and the preliminary step for each classification by the classification unit 30.

In the feature amount extraction step, the conversion unit 31 converts a plurality of classified operation data into a plurality of feature data. The conversion unit 31 extracts one or more feature quantities for each component of the operation data. When the component of the operation data is represented by a truth value, the conversion unit 31 extracts, for example, a numerical value of +1 or -1 from the true value or the false value as a feature amount. When the component of the operation data is represented by a numerical value, the conversion unit 31 extracts, for example, the numerical value as it is as a feature amount. For example, when the components of the operation data are represented by a list of numerical values in waveform data or the like, the conversion unit 31 extracts, for example, the average value and standard deviation of the numerical values included in the list as one or more feature quantities. Alternatively, when the components of the operation data are represented by a list of numerical values, the conversion unit 31 extracts, for example, a plurality of numerical values included in the list as they are as a plurality of feature quantities. The conversion unit 31 may extract a feature amount from the components of the operation data by a method not exemplified here. The conversion unit 31 generates multi-component feature data including one or more feature quantities extracted for each component of the operation data as a component.

In the standardization step, the conversion unit 31 converts a plurality of feature data into a plurality of standardized data. Standardized data is multi-component data. The conversion unit 31 converts each component of the feature data into each component of the standardized data. The components of the standardized data are standardized, for example, so that the average for the classification including the original operation data is 0. The components of the standardized data are standardized, for example, so that the standard deviation for the classification containing the original operation data is 1.

In the abnormality degree calculation step, the conversion unit 31 converts a plurality of standardized data into a plurality of abnormality degree data. The anomaly degree data is multi-component data. Each component of the anomaly degree data is an index showing the difference from the normal state. Each component of the anomaly degree data is calculated from, for example, each component of the feature data. For example, the conversion unit 31 calculates each component of the anomaly degree data by dividing the squared deviation from the mean value by the variance for each component of the feature data. The conversion unit 31 may convert the standardized data into abnormality degree data by another method such as machine learning.

In the preliminary step, the conversion unit 31 converts a plurality of abnormality degree data into a plurality of state data. The conversion unit 31 applies the unsupervised learning method to the abnormality degree data as a preliminary process. The method of unsupervised learning is, for example, a method of dimension reduction by PCA (Principal Component Analysis). Alternatively, the method of unsupervised learning is, for example, a clustering method by the k-means method. The conversion unit 31 outputs the plurality of determination data included in the training data set and the plurality of state data converted from the plurality of operation data corresponding to the plurality of determination data to the learning unit 32.

In Graph B, an example of two-component state data is shown. The horizontal axis of the graph B represents the first component of the state data. The vertical axis of the graph B represents the second component of the state data. In the graph B, each point represents the state data converted from the operation data acquired by one operation of the brake device 11. The state data may be data of one component or three or more components.

The learning unit 32 learns a diagnostic model for each classification by the classification unit 30 based on a plurality of state data, for example, by a supervised learning method using a plurality of determination data as teacher data. Here, the method of supervised learning is, for example, a linear or non-linear classification method. The method of supervised learning is, for example, k-nearest neighbor method, discriminant analysis or SVM (Support Vector Machine). At this time, the presence / absence of anomalies, the type of anomalies, and the degree of anomalies included in the determination data are labeled as clusters classified by, for example, a linear or non-linear classification method. After learning the diagnostic model, the learning unit 32 stores the diagnostic model in the built-in storage area.

In the graph B, the state data representing the normal state is represented by a white circle. According to the diagnostic model learned by the learning unit 32, the state data representing the normal state is classified into one of two clusters representing the normal state.

The determination unit 33 is activated, for example, by the operation of the operator of the information center 21. When the determination unit 33 is activated, the determination unit 33 reads the diagnostic model stored in the learning unit 32.

When the determination unit 33 is activated, when the classification unit 30 acquires the observation data from the observation unit 28, the classification unit 30 acquires the attribute data for the elevator 2 corresponding to the observation data from the attribute data storage unit 26. The classification unit 30 acquires operation data and environmental data from observation data and attribute data. The classification unit 30 classifies the operation data based on the environmental data. The classification unit 30 outputs the classified operation data to the conversion unit 31. The conversion unit 31 converts the operation data into state data according to the classification by the classification unit 30. At this time, the conversion unit 31 converts the data into state data by the same method as the operation data included in the training data set. That is, the conversion unit 31 performs conversion so that the same state data can be obtained from the same operation data. The conversion unit 31 outputs the converted state data to the determination unit 33.

The determination unit 33 determines the abnormality of the brake device 11 using the state data based on the diagnostic model read from the learning unit 32. The determination unit 33 acquires, for example, the presence / absence of an abnormality, the type of abnormality, and the degree of abnormality as a result of determining the abnormality of the brake device 11. For example, when learning is performed by a classification method such as the k-nearest neighbor method in the learning unit 32, the determination unit 33 determines whether or not there is an abnormality in the brake device 11 based on the label of the cluster in which the preprocessed data is classified. Get the type and degree of anomaly.

In the graph B, the state data representing the abnormal state is represented by a black circle. The determination unit 33 determines, for example, that the state data represents an abnormal state when the state data is not classified into either of the two clusters representing the normal state. The number of clusters representing a normal state may be one or three or more. Alternatively, the determination unit 33 may determine that the state data represents an abnormal state when the state data is classified into any one or more clusters representing an abnormal state.

The determination unit 33 outputs the determination result data to the storage unit 36. The determination result data includes the presence / absence of abnormality, the type of abnormality, and the degree of abnormality of the brake device 11 acquired by the determination unit 33 as a result of the determination. The determination result data includes abnormality degree data. The determination result data includes the data of the margin until it is determined to be abnormal. The margin until it is determined to be abnormal is, for example, the minimum value of the distance between the point corresponding to the state data in the space after dimension reduction and the center of gravity of the cluster representing the abnormality.

The storage unit 36 stores the determination result.

The determination unit 33 outputs a request signal to the control panel 12 based on the determination result. The determination unit 33 outputs a request signal as follows, for example.

The determination unit 33 determines whether or not the determination result represents an inoperable state according to a preset standard, for example, based on the presence or absence of an abnormality, the type of abnormality, or the degree of abnormality. The inoperable state is a state in which the elevator 2 cannot be operated. Alternatively, the output unit determines that the result of the determination represents the inoperable state, for example, when the state data is classified into a cluster representing an abnormality preset as an abnormality corresponding to the inoperable state. The inoperable state includes, for example, a state in which the brake device 11 does not operate, or a state in which the degree of abnormality is significantly deteriorated when the brake device 11 is operated. The determination unit 33 outputs a request signal for stopping the operation of the elevator 2 to the control panel 12 when the determination result indicates an inoperable state. The control panel 12 stops the operation of the elevator 2 according to the request signal.

The determination unit 33 determines whether or not the determination result represents an abnormal state according to a preset standard, for example, based on the presence or absence of an abnormality, the type of abnormality, or the degree of abnormality. An abnormal state is a state that is not normal. Alternatively, the determination unit 33 determines that the determination result represents an abnormal state, for example, when the state data is not classified into a cluster representing normality. Alternatively, the determination unit 33 determines that the determination result represents an abnormal state, for example, when the state data is classified into a cluster representing any of the abnormalities.

When the result of the determination indicates an abnormal state, the determination unit 33 brakes as an operation test in a state where the car 10 is stopped at the same position as the position where the car 10 was stopped when the determination was made. A request signal for operating the device 11 is output to the control panel 12. The control panel 12 causes the brake device 11 to operate as an operation test without traveling the car 10 according to the request signal. When the brake device 11 operates, the observation unit 28 acquires operation data. The classification unit 30 classifies the operation data acquired by the observation unit 28. The conversion unit 31 converts the operation data classified by the classification unit 30 into state data. The determination unit 33 again determines the abnormality of the brake device 11 based on the diagnostic model from the state data converted by the conversion unit 31. When the result of the first judgment represents an abnormal state, when it is determined that the abnormal state is similar to the state in which the operation is disabled after the operation test, for example, by referring to the abnormality history database. The determination unit 33 does not have to perform an operation test.

When the result of the re-determination represents an abnormal state, the determination unit 33 performs an operation test in a state where the car 10 is stopped at a position different from the position where the car 10 was stopped when the determination was made. A request signal for operating the brake device 11 is output to the control panel 12. The control panel 12 drives the car 10 according to the request signal. The control panel 12 causes the brake device 11 to operate as an operation test after the car 10 is stopped according to the request signal. When the brake device 11 operates, the observation unit 28 acquires operation data. The classification unit 30 classifies the operation data acquired by the observation unit 28. The conversion unit 31 converts the operation data classified by the classification unit 30 into state data. The determination unit 33 determines the abnormality of the brake device 11 based on the diagnostic model from the state data converted by the conversion unit 31.

When the result of the first determination represents an abnormal state, the determination unit 33 is a braking device in a state where the car 10 is stopped on a floor different from the floor on which the car 10 was stopped when the determination was made. The request signal for operating the eleven may be output to the control panel 12.

When determining the abnormality of the brake device 11, the determination unit 33 outputs the determination result data to the notification unit 37.

When the result of the determination of the abnormality of the brake device 11 by the determination unit 33 indicates the operable state, the notification unit 37 does not notify the determination result data. The operable state is, for example, a state in which the elevator 2 can be operated. The operable state includes, for example, a normal state or a slightly abnormal state that does not interfere with the operation.

When the result of the abnormality determination of the brake device 11 by the determination unit 33 represents a state in which the operation is not possible, the notification unit 37 generates notification data from the determination result data. The content of the notification data is, for example, the type of abnormality, the degree of abnormality, the number of state data similar to the determined state data, and after the acquisition of the similar state data in the elevator 2 from which the similar state data was acquired. Includes the time when maintenance and inspection were carried out. The notification unit 37 notifies the content of the notification data through the display device 24 by outputting the notification data to the display device 24.

The display device 24 displays the contents of the broadcast data. On the display, for example, "Abnormality of type X has occurred. The degree of abnormality is 50%. Of the 100 similar cases in the past, 50 cases will be maintained within 1 month and 70 cases will be maintained within 2 months. It is displayed. "

Subsequently, an example of the operation of the brake device abnormality diagnosis system 1 will be described with reference to FIGS. 3 and 4.
3 and 4 are flowcharts showing an example of the operation of the brake device abnormality diagnosis system according to the first embodiment.

FIG. 3 shows the operation of learning the diagnostic model of the brake device abnormality diagnosis system 1.

In step S11, the classification unit 30 acquires the training data set from the data acquisition unit 29. The classification unit 30 classifies the operation data corresponding to the environmental data based on the environmental data included in the training data set. After that, the operation of the brake device abnormality diagnosis system 1 proceeds to step S12.

In step S12, the conversion unit 31 converts the operation data into state data for each classification by the classification unit 30. After that, the operation of the brake device abnormality diagnosis system 1 proceeds to step S13.

In step S13, the learning unit 32 learns a diagnostic model of the abnormality of the brake device 11 based on the state data. After that, the operation of the brake device abnormality diagnosis system 1 proceeds to step S14.

In step S14, the learning unit 32 stores the learned diagnostic model in a storage area built in. After that, the operation of the brake device abnormality diagnosis system 1 ends.

FIG. 4 shows the operation of the brake device abnormality diagnosis system 1 for abnormality diagnosis.

In step S21, the determination unit 33 reads the diagnostic model from the learning unit 32. After that, the operation of the brake device abnormality diagnosis system 1 proceeds to step S22.

In step S22, the classification unit 30 acquires observation data from the observation unit 28. The classification unit 30 acquires attribute data from the attribute data storage unit 26. The classification unit 30 acquires operation data and environmental data from observation data and attribute data. The classification unit 30 classifies the operation data based on the environmental data. After that, the operation of the brake device abnormality diagnosis system 1 proceeds to step S23.

In step S23, the conversion unit 31 converts the operation data into state data for each classification by the classification unit 30. After that, the operation of the brake device abnormality diagnosis system 1 proceeds to step S24.

In step S24, the determination unit 33 determines the abnormality of the brake device 11 from the state data based on the read diagnostic model. After that, the operation of the brake device abnormality diagnosis system 1 proceeds to step S25.

In step S25, the determination unit 33 outputs the determination result to the notification unit 37 and the storage unit 36. After that, the operation of the brake device abnormality diagnosis system 1 proceeds to step S22.

As described above, the brake device abnormality diagnosis system 1 according to the first embodiment includes an observation unit 28, a conversion unit 31, a data acquisition unit 29, a learning unit 32, and a determination unit 33. The observation unit 28 acquires operation data regarding the operation of the brake device 11 when the brake device 11 operates. The brake device 11 brakes the car 10 of the elevator 2. The conversion unit 31 converts the operation data acquired by the observation unit 28 into state data corresponding to the failure phenomenon of the brake device 11. The data acquisition unit 29 acquires determination data for determining an abnormality in the brake device 11. The learning unit 32 learns a diagnostic model of an abnormality of the brake device 11 by a supervised learning method using the state data and the determination data. The determination unit 33 determines the abnormality of the brake device 11 based on the diagnostic model from the state data obtained by converting the operation data acquired by the observation unit 28 after the learning by the learning unit 32 by the conversion unit 31.

The determination unit 33 determines the abnormality of the brake device 11 based on the diagnostic model learned by the supervised learning method. In the method of supervised learning, the threshold value for diagnosing an abnormality is not required in advance. Therefore, the abnormality of the brake device 11 can be diagnosed based on the data whose threshold value for diagnosing the abnormality is unknown.

The observation unit 28 can acquire various data as operation data. The learning unit 32 learns by a supervised learning method. Therefore, the learning unit 32 can learn a diagnostic model based on complicated determination conditions. As a result, the determination unit 33 can distinguish between a state corresponding to a large number of types of abnormalities and a normal state. Further, the determination unit 33 can accurately detect a sudden abnormality in which the time until the abnormality occurs cannot be predicted by a single linear expression.

Further, the brake device abnormality diagnosis system 1 includes a classification unit 30. The classification unit 30 classifies the operation data based on the environmental data about the operating environment of the brake device 11. The learning unit 32 learns a diagnostic model for each classification by the classification unit 30. The determination unit 33 determines the abnormality of the brake device 11 based on the diagnostic model learned for each classification by the classification unit 30.

The learning unit 32 can perform the learning process using the data sorted into meaningful classifications. Therefore, even when a plurality of normal regions of operation data exist due to environmental factors, the learning unit 32 can learn a highly accurate diagnostic model. In addition, the amount of calculation in the learning of the learning unit 32 is reduced.

Further, the brake device abnormality diagnosis system 1 includes a notification unit 37. The notification unit 37 notifies the result of the determination of the abnormality of the brake device 11 by the determination unit 33.

The display device 24 acquires the notification data from the notification unit 37. The display device 24 displays the contents of the broadcast data. As a result, for example, the operator of the information center 21 can quickly know the result of the determination of abnormality.

Further, the brake device abnormality diagnosis system 1 includes a storage unit 36. The storage unit 36 stores the result of the determination of the abnormality of the brake device 11 by the determination unit 33. When the result of the determination indicates that the elevator 2 can be operated, the notification unit 37 does not notify the result of the determination.

The notification unit 37 does not notify the result of the determination indicating the state of low urgency. This makes it difficult for urgent information to be overlooked.

Further, when the determination result of the abnormality of the brake device 11 indicates an abnormal state, the determination unit 33 puts the car in the brake device 11 at the same position as the position of the car 10 of the elevator 2 when the determination is made. A signal for braking 10 is output. The determination unit 33 is a brake device 11 based on a diagnostic model from the state data obtained by converting the operation data acquired by the observation unit 28 when the brake device 11 brakes the car 10 at the position by the conversion unit 31. Judge an abnormality.

The determination unit 33 determines whether the result representing the abnormal state is reproduced at the same position. As a result, the determination unit 33 can determine whether the result of the previous determination is due to an accidental factor. For example, when the result representing an abnormal state is not reproduced, the determination unit 33 may determine that the result of the previous determination is due to an accidental factor. Accidental factors include, for example, electrical noise.

Further, when the determination result of the abnormality of the brake device 11 represents a state in which the abnormality is not normal, the determination unit 33 puts the car in the brake device 11 at a position different from the position of the car 10 of the elevator 2 when the determination is made. A signal for braking 10 is output. The determination unit 33 is a brake device 11 based on a diagnostic model from the state data obtained by converting the operation data acquired by the observation unit 28 when the brake device 11 brakes the car 10 at the position by the conversion unit 31. Judge an abnormality.

The determination unit 33 determines whether the result representing the abnormal state is reproduced at different positions. Thereby, the determination unit 33 can determine whether or not the result of the previous determination is due to, for example, a local abnormality of the brake drum 15 or the brake shoe 16. For example, when the result representing the abnormal state is not reproduced, the determination unit 33 may determine that the result of the previous determination is due to a local abnormality of the brake drum 15 or the brake shoe 16. For example, when the result representing the abnormal state is reproduced, the determination unit 33 may determine that the result of the previous determination is not due to a local abnormality of the brake drum 15 or the brake shoe 16.

Further, the determination unit 33 outputs a signal for stopping the operation of the elevator 2 when the result of the determination of the abnormality of the brake device 11 indicates that the elevator 2 cannot be operated.

When an abnormality occurs in which the elevator 2 cannot be operated, the elevator 2 stops immediately. This ensures the safety of the user. Further, the deterioration of the abnormality of the brake device 11 is suppressed.

Further, the learning unit 32 may learn the abnormality diagnosis model of the brake device 11 by the unsupervised learning method using the state data and the determination data.

Here, an example of abnormality diagnosis of the brake device abnormality diagnosis system 1 by a diagnostic model learned by unsupervised learning will be described with reference to FIG. Here, we will explain in detail the differences from the case where the diagnostic model is learned by supervised learning.
FIG. 5 is a diagram showing an example of abnormality diagnosis by the brake device abnormality diagnosis system according to the first embodiment.

In the graph A, an example of the data included in the operation data is shown in the same manner as in FIG.

When the brake device abnormality diagnosis system 1 starts learning, the data acquisition unit 29 generates a training data set. At this time, the training data set generated by the data acquisition unit 29 does not have to include the determination data.

The conversion unit 31 converts a plurality of operation data classified by the classification unit 30 into a plurality of state data.

In Graph C, an example of a plurality of converted state data is shown. The horizontal axis of the graph C represents the number of operations of the brake device 11. In the graph C, each point represents the state data converted from the operation data acquired by one operation of the brake device 11. The vertical axis of the graph C represents the value of the state data. The state data may be data of two or more components. In the graph C, the state data representing the normal state is represented by a white circle. In the graph C, the state data representing the abnormal state is represented by a black circle.

The learning unit 32 learns a diagnostic model for each classification by the classification unit 30 by a method of unsupervised learning based on a plurality of state data. Here, the method of unsupervised learning is, for example, a linear or non-linear classification method. The method of unsupervised learning is, for example, a method of detecting outliers. The method of unsupervised learning is, for example, one-class SVM. Alternatively, the method of unsupervised learning is, for example, the LOF (Local Outlier Factor) method. Alternatively, the unsupervised learning method is a classification method such as the k-means method. At this time, for example, if the state data is not classified into any cluster, the learning unit 32 may learn the diagnostic model by assuming that the cluster is an outlier. The learning unit 32 learns a diagnostic model so that, for example, when the state data converted from the newly acquired operation data is determined to be an outlier, the state data represents an abnormal state. The learning unit 32 may detect outliers using the data converted from the state data.

After learning the diagnostic model, the learning unit 32 stores the diagnostic model in the built-in storage area. In addition, for example, a maintenance worker may attach a label indicating the presence or absence of an abnormality, the type of abnormality, or the degree of abnormality to the cluster classified by the diagnostic model after the fact.

The determination unit 33 determines the abnormality of the brake device 11 based on the diagnostic model learned by the unsupervised learning method. In the method of unsupervised learning, the threshold value for diagnosing an abnormality is not required in advance. Therefore, the abnormality of the brake device 11 can be diagnosed based on the data whose threshold value for diagnosing the abnormality is unknown. Further, the learning unit 32 does not need the determination data.

The observation unit 28 can acquire various data as operation data. The learning unit 32 learns by a supervised learning method. Therefore, the learning unit 32 can learn a diagnostic model based on complicated determination conditions. As a result, the determination unit 33 can distinguish between a state corresponding to a large number of types of abnormalities and a normal state. Further, the determination unit 33 can accurately detect a sudden abnormality in which the time until the abnormality occurs cannot be predicted by a single linear expression.

Further, the brake device abnormality diagnosis system 1 includes a prediction unit 35 that predicts the deterioration time of the brake device 11 based on the operation data. In the abnormality diagnosis of the brake device 11, the classification unit 30 may classify the operation data based on the environmental data and the deterioration time predicted by the prediction unit 35.

In the brake device 11, different types of abnormalities may occur depending on the period of use of, for example, parts to be replaced. For example, in the early stage of the period of use, an abnormality due to a manufacturing defect or the like may occur. For example, in the middle of the period of use, an abnormality may occur due to an accidental cause. For example, at the end of the period of use, an abnormality may occur due to wear of parts or the like. Here, the prediction of the deterioration time by the prediction unit 35 reflects the usage period of the parts. The classification unit 30 classifies the operation data based on the prediction of the deterioration time by the prediction unit 35. Therefore, the learning unit 32 can learn a diagnostic model for determining an abnormality according to a usage period of a component or the like. As a result, the accuracy of the abnormality determination by the determination unit 33 is improved.

The conversion unit 31 may change the order of each step in the conversion from the operation data to the state data. The conversion unit 31 may omit one or more steps in the conversion from the operation data to the state data.

The conversion unit 31 may calculate one abnormality degree component from a plurality of components of the standardized data in the abnormality degree calculation step. Thereby, the brake device abnormality diagnosis system 1 can detect the abnormality generated in the relationship between the plurality of components of the standardized data.

Further, in the abnormality diagnosis of the brake device 11, the determination unit 33 may determine a threshold value for determining that an abnormality has occurred by using a ROC (Receiver Operating Characteristics) curve. For example, when the state data is multidimensional data, the determination unit 33 determines an abnormality using the component of the state data that maximizes the AUC (Area Under Curve) of the ROC curve. At this time, the threshold value for determining that an abnormality has occurred is, for example, a value at which Youden Index is maximized.

When the user is in the car 10, the control panel 12 may suspend the operation test until the user gets off. The result of the determination made by the determination unit 33 from the state data obtained by converting the operation data acquired by the observation unit 28 into the operation data acquired by the observation unit 28 when the brake device 11 operates while the operation test is suspended. If does not indicate an abnormal state, the control panel 12 may cancel the operation test. At this time, the determination unit 33 may add data indicating that the operation test has been canceled to the determination result data.

When it is determined that the result of the determination is due to an accidental factor, the determination unit 33 may add data indicating the possibility of being due to an accidental factor to the determination result data for the determination. Alternatively, when it is determined that the determination result is due to an accidental factor, the determination unit 33 may cancel the determination result. At this time, the determination unit 33 may modify the result of the determination as a result indicating a normal state.

When the determination result is determined to be due to, for example, a local abnormality in the brake drum 15 or the brake shoe 16, the determination unit 33 indicates the possibility that the determination result data for the determination is due to a local abnormality. Data may be added.

The notification unit 37 may notify the maintenance staff of the contents of the notification data by outputting the notification data to the maintenance terminal possessed by the maintenance staff. The notification unit 37 may notify by simultaneously outputting notification data to a plurality of output destinations.

When the elevator 2 is provided with a display device 24 for the user, for example, the notification unit 37 may notify the user of the content of the notification data by outputting the notification data to the display device 24 through the monitoring device 13 or the like. .. For example, when the content of the notification data indicates an unusable state, the display device 24 displays "This elevator cannot be used" or the like. As a result, the user can be notified of the result of the determination as to whether or not the elevator 2 can be used.

By providing the data server 22 in the information center 21, the brake device abnormality diagnosis system 1 can use the information of the elevator 2 and other elevators. As a result, the accuracy of the abnormality diagnosis of the brake device abnormality diagnosis system 1 is improved.

By providing the classification unit 30 in the information center 21, maintenance such as updating the algorithm for classifying operation data becomes easy. By providing the conversion unit 31 in the information center 21, maintenance such as updating the algorithm for converting operation data becomes easy. By providing the learning unit 32 in the information center 21, maintenance such as updating the learning algorithm of the diagnostic model becomes easy. By providing the determination unit 33 in the information center 21, maintenance such as updating the algorithm for determining an abnormality becomes easy.

The maintenance support device 23 may be provided in the building 3. At this time, the maintenance support device 23 directly communicates with, for example, the control panel 12. The maintenance support device 23 communicates with the data server 22 through, for example, the monitoring device 13. The data server 22 may be provided in the building 3.

A part or all of the functions of the brake device abnormality diagnosis system 1 may be realized in the device provided in the building 3. For example, the determination unit 33 may be realized in a device provided in the building 3. At this time, the determination unit 33 can quickly determine the abnormality of the brake device 11 without being affected by the communication failure that may occur between the building 3 and the information center 21.

The electrical connection between the systems, devices, devices, parts, etc. in Embodiment 1 may be either a direct or indirect connection. Communication of data, etc. between systems, devices, devices, parts, etc. in Embodiment 1 may be either direct or indirect communication.

Subsequently, an example of the hardware configuration of the brake device abnormality diagnosis system 1 will be described with reference to FIG.
FIG. 6 is a diagram showing a hardware configuration of a main part of the brake device abnormality diagnosis system according to the first embodiment.

Each function of the brake device abnormality diagnosis system 1 can be realized by a processing circuit. The processing circuit includes at least one processor 1b and at least one memory 1c. The processing circuit may include at least one dedicated hardware 1a with or as a substitute for the processor 1b and the memory 1c.

When the processing circuit includes the processor 1b and the memory 1c, each function of the brake device abnormality diagnosis system 1 is realized by software, firmware, or a combination of software and firmware. At least one of the software and firmware is written as a program. The program is stored in the memory 1c. The processor 1b realizes each function of the brake device abnormality diagnosis system 1 by reading and executing the program stored in the memory 1c.

The processor 1b is also referred to as a CPU (Central Processing Unit), a processing device, an arithmetic unit, a microprocessor, a microcomputer, and a DSP. The memory 1c is composed of, for example, a non-volatile or volatile semiconductor memory such as RAM, ROM, flash memory, EPROM, EEPROM, a magnetic disk, a flexible disk, an optical disk, a compact disk, a mini disk, a DVD, or the like.

When the processing circuit includes dedicated hardware 1a, the processing circuit is realized by, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination thereof.

Each function of the brake device abnormality diagnosis system 1 can be realized by a processing circuit. Alternatively, each function of the brake device abnormality diagnosis system 1 can be collectively realized by a processing circuit. For each function of the brake device abnormality diagnosis system 1, a part may be realized by the dedicated hardware 1a, and the other part may be realized by software or firmware. As described above, the processing circuit realizes each function of the brake device abnormality diagnosis system 1 by the hardware 1a, the software, the firmware, or a combination thereof.

The brake device abnormality diagnosis system according to the present invention can be applied to an elevator.

1 Brake device abnormality diagnosis system, 2 Elevator, 3 Building, 4 Hoistway, 5 Landing, 6 Landing door, 7 Hoisting machine, 8 Main rope, 9 Balanced weight, 10 Basket, 11 Brake device, 12 Control panel, 13 Monitoring device, 14 Car door, 15 Brake drum, 16 Brake shoe, 17 Coil, 18 Plunger, 19 Spring, 20 Brake control device, 21 Information center, 22 Data server, 23 Maintenance support device, 24 Display device, 25 Observation Data storage unit, 26 attribute data storage unit, 27 abnormal data storage unit, 28 observation unit, 29 data acquisition unit, 30 classification unit, 31 conversion unit, 32 learning unit, 33 judgment unit, 34 generation unit, 35 prediction unit, 36 Storage unit, 37 notification unit, 1a hardware, 1b processor, 1c memory

Claims (4)

When the brake device that brakes the car of the elevator operates, the observation unit that acquires the operation data about the operation of the brake device, and
A conversion unit that converts the operation data acquired by the observation unit into state data corresponding to the failure phenomenon of the brake device, and a conversion unit.
Learning to learn a diagnostic model of an abnormality in the brake device by an unsupervised learning method using the state data, or a judgment data for determining an abnormality in the brake device and a supervised learning method using the state data. Department and
A determination unit that determines an abnormality of the brake device based on the diagnostic model from the state data obtained by converting the operation data acquired by the observation unit after learning by the learning unit by the conversion unit.
Equipped with
When the result of the abnormality determination of the brake device represents an abnormal state, the determination unit causes the brake device to brake the car at the same position as the position of the elevator car when the determination is made. The brake device is based on the diagnostic model from the state data obtained by converting the operation data acquired by the observation unit when the brake device brakes the car at the position by outputting a signal. Judging the abnormality of
Elevator brake device abnormality diagnosis system. When the brake device that brakes the car of the elevator operates, the observation unit that acquires the operation data about the operation of the brake device, and
A conversion unit that converts the operation data acquired by the observation unit into state data corresponding to the failure phenomenon of the brake device, and a conversion unit.
Learning to learn a diagnostic model of an abnormality in the brake device by an unsupervised learning method using the state data, or a judgment data for determining an abnormality in the brake device and a supervised learning method using the state data. Department and
A determination unit that determines an abnormality of the brake device based on the diagnostic model from the state data obtained by converting the operation data acquired by the observation unit after learning by the learning unit by the conversion unit.
Equipped with
When the result of the abnormality determination of the brake device represents an abnormal state, the determination unit causes the brake device to brake the car at a position different from the position of the elevator car when the determination is made. The brake device is based on the diagnostic model from the state data obtained by converting the operation data acquired by the observation unit when the brake device brakes the car at the position by outputting a signal. Judging the abnormality of
Elevator brake device abnormality diagnosis system. The determination unit outputs a signal for stopping the operation of the elevator when the result of the determination of the abnormality of the brake device indicates that the elevator cannot be operated.
The elevator brake device abnormality diagnosis system according to claim 1 or 2 . When the brake device that brakes the car of the elevator operates, the observation unit that acquires the operation data about the operation of the brake device, and
A conversion unit that converts the operation data acquired by the observation unit into state data corresponding to the failure phenomenon of the brake device, and a conversion unit.
Learning to learn a diagnostic model of an abnormality in the brake device by an unsupervised learning method using the state data, or a judgment data for determining an abnormality in the brake device and a supervised learning method using the state data. Department and
A determination unit that determines an abnormality of the brake device based on the diagnostic model from the state data obtained by converting the operation data acquired by the observation unit after learning by the learning unit by the conversion unit.
A classification unit that classifies the operation data based on the environmental data about the operating environment of the brake device, and
A predictor that predicts the deterioration time of the brake device based on the operation data ,
Equipped with
The classification unit classifies the operation data based on the environmental data and the deterioration time predicted by the prediction unit.
The learning unit learns the diagnostic model for each classification by the classification unit.
The determination unit determines an abnormality in the brake device based on the diagnostic model learned for each classification by the classification unit.
Elevator brake device abnormality diagnosis system.

Images