Platform door diagnosis system
The platform door diagnosis system addresses the inefficiencies of existing systems by predicting motor current values based on environmental and operational parameters, ensuring accurate condition assessment and timely maintenance.
Patent Information
- Application Number
- JP2023072054
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-04-26
- Publication Date
- 2024-11-08
AI Technical Summary
Existing platform door maintenance systems fail to accurately diagnose the condition of platform doors due to variations in motor current based on installation location, time, and type, leading to inefficient maintenance and inability to grasp the door's state between visits.
A platform door diagnosis system that predicts motor current values using temperature, installation time, and operation history, compares predicted values with actual measurements, and calculates abnormality degrees to determine the door's condition, allowing for precise maintenance scheduling.
Enables accurate diagnosis of platform door conditions, predicting potential issues like grease depletion or belt slack, facilitating timely maintenance and reducing unnecessary visits.
Smart Images

Figure 2024157615000001_ABST
Abstract
Description
[Technical field]
[0001] The present invention relates to a platform door diagnostic system that diagnoses the condition of platform doors installed on platforms. [Background technology]
[0002] In recent years, the number of platforms equipped with platform doors has been increasing. In many cases, maintenance of such platform doors is performed by workers at set intervals. However, this method requires workers to visit the platform doors regardless of their condition, which is a waste if the platform doors are not in a state where maintenance is required. In addition, there is a problem in that the condition of the platform doors cannot be grasped between maintenance sessions. Therefore, technology that can be used for platform door maintenance has been studied.
[0003] Patent Document 1 describes a maintenance support system for platform doors. This maintenance support system for platform doors acquires the value of the current flowing through a motor that opens and closes the platform doors, and when another platform door is opened or closed by another motor, it determines whether or not maintenance is required by referring to the value of the current flowing through the other motor. [Prior art documents] [Patent documents]
[0004] [Patent Document 1] JP 2020-104609 A Summary of the Invention [Problem to be solved by the invention]
[0005] The technology described in Patent Document 1 judges whether or not maintenance is required when another platform door is opened or closed by another motor as described above, by referring to the current value flowing through the other motor. However, it is known that the motor current when opening and closing platform doors increases or decreases depending on the installation location on a platform, and also increases or decreases depending on the time of year. Furthermore, even at a single station, the type of platform door may differ from one platform to another. The technology described in Patent Document 1 does not take into account such increases or decreases in motor current depending on the installation location, time, and type of platform door, and may not be able to appropriately judge the state of the platform door.
[0006] Therefore, there is a demand for a platform door diagnostic system that can properly diagnose the condition of platform doors. [Means for solving the problem]
[0007] The characteristic configuration of the platform door diagnostic system according to the present invention is as follows: A platform door diagnostic system for diagnosing the state of platform doors installed on platforms, a diagnosis time setting unit for setting a diagnosis time point indicating when a diagnosis is to be performed; a prediction unit that predicts, as a current prediction value, a current value of a motor current flowing through a motor that opens and closes the platform door after the diagnosis time point based on a parameter that indicates a state of the platform door before the diagnosis time point; an actual measurement value acquisition unit that acquires, as an actual measurement value, a measurement result of a current value of the motor current at the time of the diagnosis when the platform door is opened and closed under preset conditions; an abnormality degree calculation unit that calculates an abnormality degree of the actual measurement value at the time of diagnosis based on the current prediction value and the actual measurement value; a state determination unit that determines that a state of the platform door at the time of the diagnosis is abnormal when the degree of abnormality exceeds a preset threshold value; The point is that it is equipped with the following features.
[0008] For example, when the grease of the guide bearing supporting the platform door is depleted during opening and closing of the platform door, the motor current flowing through the motor that opens and closes the platform door is larger than when the grease is not depleted. Also, when slack occurs in the belt that is provided between the motor that opens and closes the platform door and the platform door, the motor current flowing through the motor that opens and closes the platform door is larger than when slack does not occur. Therefore, with the above-mentioned configuration, it is possible to appropriately diagnose whether the state of the platform door is abnormal or not based on the current value of the motor current. Also, by diagnosing the state of the platform door based on the current value of the motor current, for example, the above-mentioned depletion of grease and slack in the belt can be predicted, and maintenance can be performed before the grease depletion and slack in the belt occur. In particular, in this configuration, the degree of abnormality is calculated by comparing the predicted current value after the diagnosis time point predicted from the parameters before the diagnosis time point with the actual measured value at the diagnosis time point, so that a diagnosis specific to the platform door can be performed that is not dependent on the installation location or installation time. Therefore, the prediction accuracy can be improved compared to the conventional case where the actual measured value obtained from another platform door is used.
[0009] In addition, it is preferable that the prediction unit uses, as the parameters, temperature information indicating the temperature at a specified time in the area where the platform doors are installed between the installation of the platform doors and the diagnosis time, and current value information including the actual measured value when the platform doors are operated to open and close in an operation confirmation test that confirms the opening and closing operation of the platform doors that was carried out under the preset conditions between the installation of the platform doors and the diagnosis time.
[0010] The motor current flowing through the motor that drives the platform doors to open and close tends to be larger when the temperature is low than when the temperature is high. Therefore, by having the prediction unit predict the motor current value flowing through the motor as a current prediction value using temperature information and current value information as in this configuration, it becomes possible to accurately predict the current value according to the temperature at the time of diagnosis.
[0011] It is also preferable that the pre-set condition is a condition in which there are no passengers on the platform.
[0012] In this way, by setting the condition that there are no passengers on the platform, it is possible to avoid a situation in which, for example, a passenger touching the platform door prevents the opening and closing of the platform door. Therefore, the actual measurement value acquisition unit can acquire measurement results of the motor current value that are not affected by passengers. Moreover, since the station staff only needs to open and close the platform door before the start of train operations, an abnormality diagnosis of the platform door is automatically performed, enabling prompt maintenance response.
[0013] In addition, it is preferable that the prediction unit creates a learned model based on the current value of the motor current when the state of the platform door is not abnormal, and predicts the current prediction value using the model.
[0014] With this configuration, the prediction accuracy of the current prediction value by the prediction unit can be improved, making it possible to more appropriately diagnose the state of the platform screen doors. In particular, by performing machine learning with teacher data using data when the platform screen doors are not abnormal, it is only necessary to extract actual values that deviate from the current prediction value by a predetermined value or more as abnormal data and determine whether there is an abnormality, and the determination speed of the state determination unit can be increased.
[0015] It is also preferable that the prediction unit also uses, as one of the parameters, the number of times the platform doors have been opened and closed since they were installed.
[0016] For example, the grease will become more familiar depending on the number of times the platform doors have been opened and closed since they were installed, and burrs on the parts will be removed, allowing the doors to move more smoothly. Therefore, by predicting the predicted current value taking into account the number of times the platform doors have been opened and closed, as in this configuration, it is possible to appropriately predict the current value of the motor current even if the current value of the motor current changes over time. This makes it possible to diagnose the condition of the platform doors with greater accuracy. [Brief description of the drawings]
[0017] [Figure 1] FIG. 2 is a block diagram showing the configuration of a platform door and a platform door diagnostic system. [Diagram 2] FIG. 4 is a diagram showing a time series change in motor current. [Diagram 3] FIG. 4 is a diagram showing a time series change in motor current. [Figure 4] FIG. 11 is a diagram showing the relationship between motor current and air temperature. [Diagram 5] FIG. 11 is a diagram showing the relationship between motor current and air temperature. [Figure 6] FIG. 11 is a diagram showing a current prediction value predicted using a past current value of a motor current. [Figure 7] FIG. 11 is a diagram showing a current prediction value predicted using a past current value of a motor current. [Figure 8] FIG. 13 is a diagram showing an abnormality degree. [Figure 9] FIG. 13 is a diagram showing an abnormality degree. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0018] The platform door diagnostic system according to the present invention is configured to be able to appropriately diagnose the state of the platform door. Hereinafter, a platform door diagnostic system 1 according to the present embodiment will be described.
[0019] Figure 1 is a block diagram showing a schematic configuration of a platform door diagnostic system 1. As shown in Figure 1, the platform door diagnostic system 1 is configured to include a diagnosis time point setting unit 11, a prediction unit 12, a parameter setting unit 13, an actual measurement value acquisition unit 14, an abnormality degree calculation unit 15, a state determination unit 16, a maintenance information storage unit 17, and a storage unit 20. Each functional unit is constructed of hardware or software, or both, with a CPU as a core component in order to perform processing related to diagnosing the state of the platform doors 3.
[0020] As shown in Fig. 1, a plurality of platform doors 3 are installed on a platform 2. Each of the plurality of platform doors 3 can be opened and closed by a motor M. The platform door diagnostic system 1 individually diagnoses the condition of each of the plurality of platform doors 3 based on the current value of the current (motor current) flowing through the motor M when the platform doors 3 are opened and closed.
[0021] Each platform door 3 is tested by station staff to check its operation every day before the first train uses the platform 2 on which the platform door 3 is installed. The current value of the motor current at this time is measured by a current sensor (not shown). The measurement results of the motor current by the current sensor are stored sequentially in the memory unit 20 together with measurement date information indicating the date of measurement for each platform door 3. The measurement results by the current sensor can be configured to be transmitted to the memory unit 20, for example, via a network.
[0022] 2 and 3 show the measurement results of the motor current stored in the memory unit 20. FIG. 2 shows the measurement results of the motor current when an opening operation is performed in an operation confirmation test of a specific platform door 3 installed on platform X at station A, and the measurement results of the motor current when a closing operation is performed. FIG. 3 shows the measurement results of the motor current when an opening operation is performed in an operation confirmation test of a specific platform door 3 installed on platform Y at station B, and the measurement results of the motor current when a closing operation is performed. In FIG. 2 and FIG. 3, the horizontal axis indicates the day when the operation confirmation test is performed, and the vertical axis indicates the current value of the motor current. As shown in FIG. 2 and FIG. 3, on the same day, the motor current when the opening operation of the platform door 3 is performed is larger than the motor current when the closing operation of the platform door 3 is performed. In addition, the motor current when the opening and closing operation of the platform door 3 is performed when the temperature is low (winter) is larger than the motor current when the opening and closing operation of the platform door 3 is performed when the temperature is high (summer). In the memory unit 20, such current values of the motor current are additionally stored each time an operation confirmation test is performed.
[0023] Returning to Fig. 1, the diagnosis time point setting unit 11 sets a diagnosis time point indicating when the diagnosis will be performed. The diagnosis time point can be set to any time point between when the motor current described above starts to be stored in the memory unit 20 and the present. In other words, the diagnosis time point does not have to be the current time point at which the diagnosis is performed, but may be a predetermined past time point between when the platform screen doors 3 were installed on the platform 2 and when the opening and closing operations were performed in the operation confirmation test and the present.
[0024] The prediction unit 12 predicts, as a current prediction value, the motor current value flowing through the motor M that opens and closes the platform doors 3 from the diagnosis time point onward, based on parameters that indicate the state of the platform doors 3 before the diagnosis time point. "Before the diagnosis time point" refers to the period from when the platform doors 3 are installed on the platform 2 to the diagnosis time point set by the diagnosis time point setting unit 11. The parameters that indicate the state of the platform doors 3 correspond to information that indicates the surrounding environment when the platform doors 3 are opened and closed, and information obtained when the platform doors 3 are opened and closed.
[0025] Here, Fig. 4 and Fig. 5 show the relationship between the motor current and the temperature. Fig. 4 shows the relationship between the motor current and the temperature when the opening operation is performed in the operation confirmation test of a specific platform door 3 installed on platform X of station A, and the relationship between the motor current and the temperature when the closing operation is performed. Also, Fig. 5 shows the relationship between the motor current and the temperature when the opening operation is performed in the operation confirmation test of a specific platform door 3 installed on platform Y of station B, and the relationship between the motor current and the temperature when the closing operation is performed. In Fig. 4 and Fig. 5, the horizontal axis indicates the temperature when the opening and closing operation of the platform door 3 is performed, and the vertical axis indicates the current value of the motor current. As shown in Fig. 4 and Fig. 5, in the case of the same temperature, the motor current when the opening operation of the platform door 3 is performed tends to be larger than the motor current when the closing operation of the platform door 3 is performed. Also, Fig. 4 and Fig. 5 show that the motor current when the opening and closing operation of the platform door 3 is performed when the temperature is low as described above is larger than the motor current when the opening and closing operation of the platform door 3 is performed when the temperature is high.
[0026] Therefore, it has been found that when predicting the current value of the motor current, it is effective to use temperature information indicating the temperature at a predetermined time in the area where the platform doors 3 are installed between the installation of the platform doors 3 and the diagnosis time as information indicating the surrounding environment when the platform doors 3 are opened and closed. That is, it is effective to use information indicating the temperature measured at a predetermined time in the place where the platform doors 3 are installed. Specifically, it is effective to use information indicating the temperature measured at the place where the platform doors 3 are installed at a fixed time at least once a day from the installation of the platform doors 3 to the diagnosis time. Such temperature information may be measured and obtained by station staff at the station where the platform doors 3 are installed, for example, or may be measured and obtained automatically. Of course, it may also be obtained from weather information transmitted via the Internet.
[0027] The information obtained when the platform door 3 is opened and closed corresponds to current value information including the actual measured value of the motor current when the platform door 3 is opened and closed in an operation confirmation test for confirming the opening and closing operation of the platform door 3 performed under preset conditions from the time the platform door 3 is installed until the time of diagnosis. Here, as described above, an operation confirmation test of the platform door 3 is performed by station staff every day before the first train uses the platform 2. Therefore, the preset condition corresponds to the time before the first train of the day at the platform 2, and means a condition in which there are no passengers on the platform 2. The current value of the motor current is the current value of the current flowing through the motor M that drives the platform door 3 to open and close. In this embodiment, the current value information is measured by a current sensor (not shown) when the operation confirmation test is performed and stored in the memory unit 20.
[0028] Therefore, the prediction unit 12 uses, as parameters, temperature information indicating the temperature at a predetermined time in the area where the platform doors 3 are installed between the installation of the platform doors 3 and the diagnosis time, and current value information including the actual measured value when the platform doors 3 are driven to open and close in an operation confirmation test that confirms the opening and closing operation of the platform doors 3 carried out under conditions between the installation of the platform doors 3 and the diagnosis time. Note that the above-mentioned parameters are only examples, and other information can be used as parameters. Furthermore, such parameters can be set by the parameter setting unit 13 according to the intention of the diagnoser or automatically in response to the diagnosis of the platform doors 3.
[0029] The prediction unit 12 can create a model learned based on the current value of the motor current when the state of the platform doors 3 is not abnormal, and predict the predicted current value using this model. That is, the platform doors 3 are opened and closed in a state in which it is known in advance that the state of the platform doors 3 is not abnormal, and the prediction unit 12 is made to perform machine learning to learn the current value of the motor current acquired at that time as the current value when there is no abnormality, and the current value is predicted using the model thus machine-learned. This makes it possible to obtain the predicted current value of the motor current.
[0030] The prediction unit 12 can also predict the current value of the motor current flowing through the motor M that opens and closes the platform doors 3, for example, by using a prediction model according to Non-Patent Document 1, cited below. This prediction model can be implemented with "outlier detection" that detects abnormal values that occur temporarily, "abnormal part detection" that detects waveforms that differ from normal in periodic data, and "change point detection" that detects changes in measured values that occur at a certain point in time. In this embodiment, it is preferable to use "outlier detection" or "change point detection", and the following description will use "outlier detection". [Non-patent document 1] “PROPHET Forecasting at scale”, [online], [searched on March 28, 2021], Internet<URL:https: / / facebook.github.io / prophet / >
[0031] 6 and 7 show the prediction results of the motor current value flowing through the motor M when the platform door 3 is opened and closed using such a prediction model. FIG. 6 shows the motor current value predicted from October 1, 2020 onwards, using the motor current value results from February 16, 2017 to September 30, 2020 obtained during the operation confirmation test of a specific platform door 3 installed on platform X at station A. Note that FIG. 6 also shows the motor current value obtained during the operation confirmation test after October 1, 2020 for reference. Also, FIG. 7 shows the motor current value predicted from April 1, 2021 onwards, using the motor current value results from March 7, 2019 to March 31, 2021 obtained during the operation confirmation test of a specific platform door 3 installed on platform Y at station B. Note that FIG. 7 also shows the motor current value obtained during the operation confirmation test after October 1, 2020 for reference.
[0032] From Figures 6 and 7, the motor current value at the time of diagnosis predicted based on the motor current value obtained before the diagnosis time and the current value after the diagnosis time are comparable to the motor current value actually obtained during the operation check test, and the tendency is also similar. On the other hand, for example, the predicted value marked K1 in Figure 6 and the predicted value marked K2 in Figure 7 have a larger deviation from the motor current value actually obtained during the operation check test compared to other parts. This is thought to be due to the fact that maintenance of the platform door 3 (for example, application of grease to the moving parts of the platform door 3) was performed on the day with the predicted value K1 or the day with the predicted value K2.
[0033] In the present platform door diagnostic system 1, the predicted value of the motor current predicted by the prediction unit 12 as described above is compared with the actual measured value of the motor current obtained when the platform door 3 is actually opened and closed, and whether or not the state of the platform door 3 is abnormal is determined based on the actual measured value. Then, the actual measured value acquisition unit 14 acquires, as the actual measured value, the measurement result of the motor current at the time of diagnosis when the platform door 3 is opened and closed under a preset condition. In this embodiment, the preset condition is a condition in which there are no passengers on the platform 2, and specifically, corresponds to the time before the first train of the day at the platform 2. The operation confirmation test for confirming the opening and closing operation of the platform door 3 is a test for confirming the opening and closing operation of the platform door 3, which is performed by a station staff member before the above-mentioned first train. As described above, the motor current, which is the current flowing through the motor M at this time, is stored sequentially in the storage unit 20. Therefore, the actual measurement value acquisition unit 14 acquires, as an actual measurement value, the current value of the motor current at the diagnosis time point set by the diagnosis time point setting unit 11 from the memory unit 20, which stores the motor current during the test to check the opening and closing operation of the platform doors 3 performed by station staff before the first train of the day at platform 2.
[0034] The anomaly degree calculation unit 15 calculates the degree of anomaly of the actual measurement value at the time of diagnosis based on the current predicted value and the actual measurement value. The current predicted value is transmitted from the prediction unit 12, and the actual measurement value is transmitted from the actual measurement value acquisition unit 14. The degree of anomaly indicates the degree of abnormality of the actual measurement value at the time of diagnosis, and in this embodiment is expressed by the following formula (1).
number
[0035] 8 and 9 show the calculation results of the degree of abnormality calculated by the abnormality degree calculation unit 15. Fig. 8 shows the calculation result of the degree of abnormality after the diagnosis of a specific platform door 3 installed on platform X of station A, which is shown as the "predicted range" in Fig. 6. Also, Fig. 9 shows the calculation result of the degree of abnormality after the diagnosis of a specific platform door 3 installed on platform Y of station B, which is shown as the "predicted range" in Fig. 7. Such calculation results of the degree of abnormality by the abnormality degree calculation unit 15 are transmitted to the state determination unit 16, which will be described later.
[0036] When the degree of abnormality exceeds a preset threshold, the state determination unit 16 determines that the state of the platform door 3 at the time of diagnosis is abnormal. In this embodiment, as shown in Figs. 8 and 9, the threshold used for determining the state of the platform door 3 is set to a part where the degree of abnormality is 0.10. This threshold may be set by taking into consideration the overall tendency of the predicted value or the actual measurement value, or may be set to a fixed value in advance. When the degree of abnormality exceeds such a threshold, the state determination unit 16 may determine that the state of the platform door 3 is abnormal at the time of diagnosis (the time when the threshold is exceeded). For example, the degree of abnormality marked with A1 in Fig. 8 exceeds the threshold, and the degree of abnormality marked with A2 in Fig. 9 exceeds the threshold. Therefore, the state determination unit 16 may normally determine that the state of the platform door 3 is abnormal on a day corresponding to the degree of abnormality marked with A1 in Fig. 8 or on a day corresponding to the degree of abnormality marked with A2 in Fig. 9.
[0037] On the other hand, as described above, it is known that when maintenance is performed on the platform door 3, the current value (actual measured value) of the motor current becomes larger than the predicted value of the motor current. Therefore, when maintenance is performed on the platform door 3, the degree of abnormality becomes larger. Therefore, when determining whether the state of the platform door 3 is abnormal, the state determination unit 16 acquires maintenance information from the maintenance information storage unit 17, which stores maintenance information indicating the date and content of maintenance performed on the platform door 3, and if maintenance is performed on the platform door 3 between the diagnosis time and a preset time before, it is preferable not to determine that the state of the platform door 3 is abnormal even if the degree of abnormality exceeds the threshold. For example, if maintenance is performed on a day corresponding to the degree of abnormality marked with A1 in FIG. 8 and a day corresponding to the degree of abnormality marked with A2 in FIG. 9, the state determination unit 16 does not determine that the state of the platform door 3 is abnormal based on the degree of abnormality marked with A1 and the degree of abnormality marked with A2. This makes it possible to prevent erroneous determination that the platform door 3 is abnormal because the current value of the motor current has increased due to maintenance.
[0038] Other embodiments In the above embodiment, it has been explained that the status determination unit 16 does not determine that the status of the platform doors 3 is abnormal even if the degree of abnormality exceeds the threshold value if maintenance is performed on the platform doors 3 between the time of diagnosis and a preset time ago. Alternatively, the status determination unit 16 can be configured to, for example, notify or clearly display a message requesting checking of the platform doors 3 when the degree of abnormality exceeds the threshold value if maintenance is performed on the platform doors 3 between the time of diagnosis and a preset time ago.
[0039] In the above embodiment, it has been described that the prediction unit 12 is capable of creating a model learned based on the current value of the motor current when the state of the platform door 3 is not abnormal, predicting the current prediction value using the model, and further predicting the current prediction value by using "PROPHET Forecasting at scale". However, the prediction unit 12 can also be configured to predict the current prediction value by a method other than these. In other words, the prediction model is not limited to the above prediction model, and a known algorithm capable of outputting a current prediction value using the current value of the motor current when the state of the platform door 3 is not abnormal as input data may be used.
[0040] In the above embodiment, the prediction unit 12 has been described as using, as parameters, temperature information indicating the temperature at a predetermined time between the installation of the platform doors 3 in the area where the platform doors 3 are installed and the diagnosis time, and current value information including the actual measured value when the platform doors 3 are opened and closed in an operation confirmation test for confirming the opening and closing operation of the platform doors 3 performed under conditions between the installation of the platform doors 3 and the diagnosis time. For example, the prediction unit 12 can also be configured to use, as a parameter, the number of times the platform doors 3 have been opened and closed since their installation. Since the number of times the platform doors 3 have been opened and closed since their installation generally correlates with the maintenance period of the platform doors 3, this makes it possible to improve the prediction accuracy of the prediction unit 12. In addition, the parameter is not limited to temperature, and may be meteorological information such as wind speed, or a combination of multiple other meteorological information (wind speed, etc.) in addition to temperature may be used.
[0041] In the above embodiment, the preset condition is a condition that there are no passengers on platform 2, and a specific example has been described in which there is before the first train of the day on platform 2. However, the preset condition may be a condition that there are no passengers on platform 2, after the last train, or the like, or a condition that there are very few passengers.
[0042] In the above embodiment, prediction results and calculation results of the degree of abnormality are shown not only for a predetermined diagnosis time point but also for a predetermined range (several months) in relation to the prediction by the prediction unit 12 and the calculation of the degree of abnormality by the degree of abnormality calculation unit 15. However, the prediction by the prediction unit 12 and the calculation of the degree of abnormality by the degree of abnormality calculation unit 15 can also be the prediction result and the calculation result of the degree of abnormality for a predetermined day. [Industrial Applicability]
[0043] The present invention can be used in a platform door diagnostic system that diagnoses the condition of platform doors installed on platforms. [Explanation of symbols]
[0044] 1: Platform door diagnostic system 2: Platform 3: Platform doors 11: Diagnosis time setting section 12: Prediction section 14: Actual measurement acquisition section 15: Anomaly degree calculation unit 16: Status determination unit M: Motor
Claims
1. A platform door diagnostic system for diagnosing the state of platform doors installed on platforms, a diagnosis time setting unit for setting a diagnosis time point indicating when a diagnosis is to be performed; a prediction unit that predicts, as a current prediction value, a current value of a motor current flowing through a motor that opens and closes the platform door after the diagnosis time point based on a parameter that indicates a state of the platform door before the diagnosis time point; an actual measurement value acquisition unit that acquires, as an actual measurement value, a measurement result of a current value of the motor current at the time of the diagnosis when the platform door is opened and closed under preset conditions; an abnormality degree calculation unit that calculates an abnormality degree of the actual measurement value at the time of diagnosis based on the current prediction value and the actual measurement value; a state determination unit that determines that a state of the platform door at the time of the diagnosis is abnormal when the degree of abnormality exceeds a preset threshold value; A platform door diagnostic system equipped with
2. The platform door diagnosis system of claim 1, wherein the prediction unit uses, as the parameters, temperature information indicating the temperature in the area where the platform doors are installed at a specified time between the installation of the platform doors and the diagnosis time, and current value information including the actual measured value when the platform doors are operated to open and close in an operation confirmation test that confirms the opening and closing operation of the platform doors that was carried out under the preset conditions between the installation of the platform doors and the diagnosis time.
3. 3. The platform door diagnostic system according to claim 1 or 2, wherein the preset condition is a condition in which there are no passengers on the platform.
4. The platform door diagnostic system of claim 1 or 2, wherein the prediction unit creates a learned model based on the current value of the motor current when the state of the platform door is not abnormal, and predicts the current predicted value using the model.
5. The platform door diagnostic system according to claim 2 , wherein the prediction unit also uses, as one of the parameters, the number of times the platform door has been opened and closed since it was installed.
Citation Information
Patent Citations
Platform door maintenance support system, platform door and platform door maintenance support method
JP2020104609A
