Method for supporting maintenance, maintenance supporting system, maintenance supporting device, and computer program

Abstract

To provide a method for supporting maintenance, a maintenance supporting system, a maintenance supporting device, and a computer program which can continuously and stably operate a system including a power storage element.SOLUTION: The method for supporting maintenance includes the steps of: detecting a sign of an abnormality in a power storage element on the basis of measurement data on the power storage element sequentially stored in a storage device; determining at least one of the construction time and the number of operators of a maintenance operation corresponding to the sign of the detected abnormality, and articles including exchangeable articles and tools necessary for the maintenance work; and notifying the operators of determined issues and execution of the corresponding maintenance operation.SELECTED DRAWING: Figure 8

Description

[Technical field]

[0001] The present invention relates to a maintenance support method, a maintenance support system, a maintenance support device, and a computer program for supporting maintenance management work for energy storage elements. [Background technology]

[0002] Energy storage elements are widely used in devices connected to renewable energy sources such as solar power generation and wind power generation, uninterruptible power supplies, and DC or AC power supplies included in stabilized power supplies. Energy storage elements are used as backup power sources in the event of a power supply problem from a power generation system. Since power supply problems cause great disruption to social activities, it is extremely important to stabilize the power supply using energy storage elements.

[0003] Energy storage elements used as backup power sources are in a standby state when the power supply from the power generation system is normal, i.e., when no disasters or power system problems have occurred. Even if the energy storage elements are in a standby state and are not being charged or discharged, they gradually deteriorate from the time of manufacture, and there is a possibility that they will not be able to charge or discharge properly when they are actually required to be operated. Depending on the individual characteristics of the energy storage elements and the environment in which they are used, they may not be able to perform as expected at the time of manufacture.

[0004] Patent Document 1 discloses a monitoring and control device that, when an abnormality occurs in a power system, automatically creates a report including information that captures the progress, reducing the burden on the power system operator, and enables the report to be verified and appropriate measures to be taken. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2010-206985 A Summary of the Invention [Problem to be solved by the invention]

[0006] Preventive maintenance is essential for energy storage elements, such as performing maintenance inspections that include observing the progress of deterioration, and replacing them appropriately if the deterioration is more advanced than expected. If an abnormality is found in an energy storage element during maintenance inspection, or if an abnormality is automatically detected by a monitoring and control device, the energy storage element will have to be shut down during the process of verifying the abnormality, confirming with the customer that repairs will be carried out based on the verification results, arranging for the repairs, and actually carrying out the repair work to return the element to a normal state. In order not to disrupt social activities, the power supply must be stable 24 hours a day, 365 days a year, and the shutdown period of energy storage elements in standby mode must be kept as short as possible, and if possible, to zero.

[0007] To shorten downtime, frequent maintenance inspections are expected to prevent problems before they occur. However, frequent maintenance inspections require a large amount of human resources, making them difficult to implement. There is a need to reduce the burden on maintenance workers, achieve efficient and reliable maintenance of systems that include energy storage elements, and ensure a stable supply of power even when unpredictable situations occur.

[0008] An object of the present invention is to provide a maintenance support method, a maintenance support system, a maintenance support device, and a computer program for stably operating a system including an energy storage element. [Means for solving the problem]

[0009] The maintenance support method detects signs of abnormality in a storage element based on measurement data on the storage element that is sequentially stored in a memory device, determines at least one of the duration of maintenance work corresponding to the detected signs of abnormality, the number of workers, and items including replacement parts or tools required for the maintenance work, and notifies the maintenance workers of the determined items and the implementation of the maintenance work corresponding to the determined items. [Brief description of the drawings]

[0010] [Figure 1] An overview of the maintenance support system is shown below. [Diagram 2] 2 is a block diagram showing an internal configuration of a device included in the maintenance support system. FIG. [Diagram 3] FIG. 2 is a block diagram showing an internal configuration of a maintenance device. [Figure 4] 13 is a flowchart illustrating an example of a processing procedure for detecting a sign of abnormality in a second phase in the remote monitoring system. [Diagram 5] An overview of the decision model is shown below. [Figure 6] 13 is a flowchart showing another example of the processing procedure for detecting a sign of abnormality in the second phase in the remote monitoring system. [Figure 7] An overview of the image judgment model is shown below. [Figure 8] 13 is a flowchart showing an example of a processing procedure in a third phase in the maintenance support system. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] The maintenance support method detects signs of abnormality in a storage element based on measurement data on the storage element that is sequentially stored in a memory device, determines at least one of the duration of maintenance work corresponding to the detected signs of abnormality, the number of workers, and items including replacement parts or tools required for the maintenance work, and notifies the maintenance workers of the determined items and the implementation of the maintenance work corresponding to the determined items.

[0012] With the above configuration, signs of abnormality are automatically detected based on the stored measurement data, reducing the burden on maintenance workers. By sequentially storing the measurement data, the measurement data can be used by a computer to detect signs of abnormality at the early stage and determine the necessary maintenance work. Compared to a case where a maintenance worker actually inspects the storage element after an abnormality appears and investigates the cause, the time required to investigate the cause can be shortened, and the downtime of the device including the storage element can be shortened.

[0013] When the energy storage element includes a plurality of energy storage elements, the detection of the abnormality sign is performed for each of the plurality of energy storage elements or for a group of the energy storage elements. For the energy storage element or energy storage element group for which a sign has been detected, a construction period, the number of workers or items, and the workers and work date and time are determined so that the operation of the other energy storage elements can be continued.

[0014] When multiple storage elements are included, it is possible to take measures to continue operation as a whole even if some of the storage elements have an abnormality. With the above configuration, operation continues using the other storage elements that have not detected any signs of abnormality, and a stable state can be maintained 24 hours a day, 365 days a year.

[0015] A judgment model may be used that is trained to output a score corresponding to whether or not the measurement data includes measurement data of a heterogeneous storage element when measurement data measured for each storage element or for each storage element group including a plurality of the storage elements is input. When it is determined that the measurement data includes measurement data of a heterogeneous storage element based on the score output by the judgment model, heterogeneity is determined, and a sign of an abnormality in the storage element is detected based on the determined heterogeneity.

[0016] The above configuration enables more accurate and rapid detection of abnormal signs than manual analysis of measurement data. By using heterogeneity and treating heterogeneous energy storage elements that do not match the assumed model at the time of manufacture as the cause of an abnormality, it is expected that the energy storage elements as a whole will match the assumed model, improving the accuracy of abnormality detection and life expectancy prediction.

[0017] The maintenance support method may include a process of creating a time distribution of scores output from the judgment model in response to input of measurement data of the storage element group, and distinguishing urgency and determining heterogeneity based on the created time distribution.

[0018] With the above configuration, the heterogeneity of the energy storage elements can be determined by distinguishing whether it is urgent or not according to the change in the heterogeneity score over time, and maintenance work can be efficiently carried out according to the urgency.

[0019] The maintenance support method uses an image judgment model that is trained to output a score corresponding to whether or not measurement data of a storage element or a storage element group corresponding to a time distribution includes measurement data of a heterogeneous storage element when an image of the created time distribution is input. The maintenance support method may include processing of imaging the created time distribution and inputting the imaging data to the image judgment model, and determining heterogeneity of the storage element group based on the score output from the image judgment model.

[0020] The above configuration enables more accurate and rapid detection of abnormal signs and identification of heterogeneity than manual analysis of measurement data. Being able to identify heterogeneity enables rapid maintenance work based on accurate preparation.

[0021] The maintenance support method may include a process of determining a worker and a date and time for performing the maintenance work based on a list of workers who can perform the maintenance work and schedule data for each of the workers, and notifying the determined worker of the performance of the maintenance work.

[0022] With the above configuration, workers who can perform the work within the required work period are automatically determined based on the schedule data, realizing the allocation of work according to an appropriate schedule.

[0023] The maintenance support method may include receiving approval for carrying out the maintenance work from an owner of the energy storage element, and notifying the owner of the maintenance work when the approval for carrying out the maintenance work is received.

[0024] This allows maintenance work to be carried out smoothly with the consent of the owner of the energy storage element. Depending on the nature of the abnormality in the energy storage element or the nature of the maintenance work, the owner can choose not to consent.

[0025] The maintenance support method may include, when the approval for implementation is received, determining an implementation date and time based on operation information of the device obtained from the owner.

[0026] Since the actual date and time of implementation is decided based on the operation information of the device, the owner can select whether or not the maintenance work can be implemented while continuing to operate the device including the energy storage element.

[0027] If the approval is received, an order may be automatically placed for the item required for the maintenance work to a vendor of the item. Identification data of the ordered item may be notified to the worker.

[0028] With the above configuration, ordering work is also performed automatically, reducing the burden on maintenance workers or sales staff.

[0029] Estimate data for the maintenance work may be created based on the determined construction period, number of workers, and necessary items, and approval for implementation may be received based on the estimate data.

[0030] With the above-mentioned configuration, the burden on the maintenance worker or sales person in creating an estimate can be reduced, and meetings with the customer regarding the implementation of maintenance and inspection can be carried out smoothly.

[0031] The storage element may be a storage element provided in an uninterruptible power supply. With the above configuration, maintenance and inspection for preventing trouble in the storage element used as a backup power source can be efficiently and reliably performed. This contributes to stabilizing the power supply. Uninterruptible power supplies are used for backup during power outages. Since it is unacceptable that backup cannot be provided when a power outage occurs, it is very important to detect signs of abnormality to prevent trouble.

[0032] The maintenance support method can be implemented by a system including a plurality of devices. The maintenance support system includes a storage device that periodically acquires and sequentially stores measurement data related to the storage element, a maintenance terminal device that can be connected to the storage device, and a maintenance support device that can be connected for communication from the maintenance terminal device. When an abnormality or a sign of an abnormality is detected in the storage element based on the measurement data related to the storage element, the maintenance support device determines at least one of the period of maintenance work related to the detected abnormality or sign of an abnormality, the number of workers, and items including replacement items or tools required for the maintenance work. The maintenance support device transmits an implementation instruction including the determined items to the worker performing the maintenance work.

[0033] The maintenance support device includes a decision unit that, when a sign of an abnormality is detected in the storage element based on measurement data regarding the storage element that is sequentially stored in a memory device in association with identification data that identifies the storage element, decides at least one of the duration of maintenance work related to the detected sign of abnormality, the number of workers, and items including replacement items or tools required for the maintenance work, and a transmission unit that transmits implementation instructions including the decided items to the worker performing the maintenance work.

[0034] The maintenance support method may be realized as a computer program, which causes a computer to execute a process of detecting a sign of abnormality in an energy storage element based on measurement data related to the energy storage element sequentially stored in a storage device in association with identification data for identifying the energy storage element, determining at least one of a construction period for maintenance work corresponding to the detected sign of abnormality, the number of workers, and items including replacement parts or tools required for the maintenance work, and notifying the maintenance workers of the determined items and the implementation of the maintenance work corresponding to the determined items.

[0035] The present invention will now be described in detail with reference to the drawings showing embodiments thereof.

[0036] FIG. 1 shows an overview of a maintenance support system 100. The maintenance support system 100 includes a maintenance support device 1 and a maintenance terminal device 2 used by a maintenance worker. The maintenance support system 100 is communicatively connected to a remote monitoring system 300. The maintenance support system 100 is communicatively connected to a customer data management system 400. In this embodiment, the maintenance support system 100, the remote monitoring system 300, and the customer data management system 400 are managed by the manufacturer of the energy storage element 50 to be maintained, and are communicatively connected to each other via a network MN for the manufacturer or a dedicated line. The maintenance support system 100 may be communicatively connected to a manufacturing management system (not shown) for the energy storage element 50.

[0037] The network MN is a local network for the manufacturer. The network MN may be, for example, Ethernet (registered trademark) or an optical fiber line. The network MN may include a Virtual Private Network (VPN) and connect the systems 100, 300, and 400 in different locations as a local network. The lines between the maintenance support system 100 and the remote monitoring system 300 and between the maintenance support system 100 and the customer data management system 400 may be part of the network MN, or may be dedicated lines or VPNs.

[0038] The maintenance terminal device 2 and the maintenance support device 1 can be connected to each other through a communication network N or a network MN. The communication network N is the so-called Internet. The communication network N may include a carrier network that realizes wireless communication according to a predetermined mobile communication standard. The communication network N may include a general optical line.

[0039] The customer data management system 400 stores data on customers who have purchased the energy storage element to be maintained. The customer data management system 400 stores attribute data such as the customer's name or title, customer contact information, address, etc., in association with a customer ID. When a customer installs and manages multiple energy storage devices 5 at different locations, the customer data management system 400 stores the location in association with a location ID that identifies the location. The customer data management system 400 stores the serial number of the energy storage element 50 purchased by the customer in association with the customer ID. When a customer installs and manages multiple energy storage devices 5 at different locations, the customer data management system 400 stores the serial number of the installed energy storage element 50 in association with the customer ID and location ID.

[0040] The remote monitoring system 300 collects data indicating the state of the storage element 50 to be maintained, and enables remote viewing of the state based on the data collected via a network. The remote monitoring system 300 sequentially stores the state data of the storage element 50 in association with the serial number of the storage element 50. When the remote monitoring system 300 receives state data of the storage element 50, it determines whether or not there is a sign of abnormality in the storage element 50 using a determination model 3M that outputs a score related to a sign of abnormality. The remote monitoring system 300 may derive diagnostic data including a state of charge (SOC), a state of health (SOH), and a predicted lifespan for each storage element 50 based on the state data.

[0041] The manufacturing management system may store the manufacturing lot number and shipping date and time in association with the manufacturing number of the energy storage element 50.

[0042] The energy storage elements 50 to be maintained by the maintenance support system 100 are preferably rechargeable, such as secondary batteries including lead-acid batteries and lithium-ion batteries, or capacitors. Some of the energy storage elements 50 may be non-rechargeable primary batteries. Each of the energy storage elements 50 in this embodiment is a lead-acid battery. The energy storage elements 50 may be energy storage modules in which a plurality of energy storage cells are connected. The energy storage elements 50 may be the energy storage cells themselves, or a group of energy storage modules in which a plurality of energy storage modules are connected.

[0043] The power storage device 5 includes one or more storage elements 50. The power storage device 5 may be used alone. The power storage device 5 may be used as a group of power storage devices 5 that are communicatively connected to a customer network CN managed by a customer (user) of the power storage element 50. The group of power storage devices 5 managed by the same customer transmits status data of the power storage element 50 to a management device 51 managed by the customer via the customer network CN. The status data includes at least a voltage value, and may also include an internal resistance value, a current value, and a temperature. The status data is transmitted from a unit connected to a terminal of the power storage element 50, which is a lead-acid battery, to the management device 51 via a maintenance communication device 6. The status data may be transmitted by a maintenance communication device 6 connected to a battery management unit (BMU) provided in a power storage module including a lithium-ion battery. The status data may be transmitted from the maintenance communication device 6 to the maintenance terminal device 2. The status data transmitted from the multiple power storage devices 5 is transmitted to the remote monitoring system 300 via a dedicated line N2 or a communication network N. The status data is stored as a status history in association with identification data, such as a serial number, that identifies each energy storage element 50. The identification data that identifies each energy storage element 50 is stored for each energy storage device 5 in association with the identification data that identifies the energy storage device 5.

[0044] A maintenance communication device 6 is provided in the power storage device 5. The maintenance communication device 6 can exchange data with a maintenance terminal device 2 used by a maintenance worker without going through a network CN. The maintenance communication device 6 can be communicatively connected to a unit that acquires status data for each of the storage elements 50 of the power storage device 5. The maintenance communication device 6 in this embodiment can be communicatively connected to a unit connected to a terminal of a lead-acid battery by wireless communication. The maintenance communication device 6 may be communicatively connected to a battery management unit (BMU) provided in a power storage module of a lithium-ion battery. The maintenance communication device 6 stores in a built-in memory the same status data as the status data transmitted from the power storage device 5 to the management device 51.

[0045] The network CN is a local network of a customer who operates multiple power storage devices 5. The network CN is, for example, Ethernet (registered trademark), and may be an optical fiber line. The network CN may include a VPN. The network CN may be an ECHONET (registered trademark) / ECHONETLite (registered trademark) compatible network. The dedicated line N2 is a private network that connects between the customer of the power storage device 5 and the remote monitoring system 300. The dedicated line N2 may be a communication network N. The dedicated line N2 may be an ECHONET / ECHONETLite compatible dedicated network.

[0046] The maintenance support system 100 of this embodiment supports the maintenance management of the energy storage element 50 or the energy storage device 5. To support the maintenance management, the maintenance support system 100 uses the status data acquired via the maintenance terminal device 2 or the customer's network CN, the customer data acquired from the customer data management system 400, and the diagnosis data acquired from the remote monitoring system 300. The maintenance support system 100 detects signs of abnormality from the aggregated status data in the remote monitoring system 300, and the maintenance support device 1 supports the work before the occurrence of a problem based on the detected abnormality or signs of abnormality. The maintenance support device 1 proceeds with arrangements for maintenance work to prevent the problem from occurring in accordance with the contents of the signs of abnormality detected, and after receiving confirmation from the customer, stops the customer's system in a planned manner to carry out the maintenance work. There is no need for the maintenance worker to manually analyze the status data. Since the maintenance support device 1 deals with the problem before it occurs, the maintenance work can be carried out according to a schedule planned in advance so as not to interfere with operation, and as a result, the downtime of the customer's system can be shortened. When a plurality of energy storage elements 50 are used to operate the energy storage device 5, the maintenance support device 1 can detect a sign of abnormality in some of them and plan maintenance work to continue the operation of the energy storage device 5. The maintenance worker can simply perform the prearranged maintenance work according to the preplanned schedule, and therefore does not need to visit the customer's system multiple times.

[0047] A detailed configuration for realizing such a maintenance support system 100 for the energy storage elements 50 will be described.

[0048] 2 is a block diagram showing the internal configuration of the devices included in the maintenance support system 100. The maintenance support device 1 uses a server computer and includes a control unit 10, a storage unit 11, and a communication unit 12. In this embodiment, the maintenance support device 1 is described as one server computer, but the processing may be distributed among multiple server computers.

[0049] The control unit 10 is a processor using a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit), and uses built-in memories such as ROM and RAM to control each component to execute processing. The control unit 10 executes processing based on a maintenance support program 1P stored in the storage unit 21.

[0050] The storage unit 11 uses a non-volatile memory such as a hard disk or an SSD (Solid State Drive). The storage unit 11 stores the above-mentioned maintenance support program 1P. The maintenance support program 1P may be a maintenance support program 7P stored in the recording medium 7 that is read by the control unit 10 and copied into the storage unit 11. The storage unit 11 stores worker data including the worker ID of the maintenance worker. The worker data includes contact information such as the worker name and email address associated with the worker ID.

[0051] The communication unit 12 is a communication device that realizes communication connection and data transmission / reception via the network MN. Specifically, the communication unit 12 is a network card compatible with the network MN. The communication unit 12 may realize communication via the communication network N via a router device (not shown) connected to the network MN. The control unit 10 transmits and receives data between the remote monitoring system 300 and the customer data management system 400 via the communication unit 12.

[0052] The maintenance terminal device 2 is a computer used by a maintenance worker. The maintenance terminal device 2 may be a desktop or laptop personal computer, or may be a so-called smartphone or tablet communication terminal. The maintenance terminal device 2 includes a control unit 20, a storage unit 21, a first communication unit 22, a second communication unit 23, a display unit 24, and an operation unit 25. The maintenance terminal device 2 may include an imaging unit 26 as shown in the figure.

[0053] The control unit 20 is a processor using a CPU or a GPU. The control unit 20 displays a repair procedure on the display unit 24 based on the maintenance terminal program 2P stored in the storage unit 21. The control unit 20 executes a process of reading information data from the maintenance communication device 6. The control unit 20 executes information processing between the maintenance support device 1 and the maintenance support device 1 using a Web browser included in the maintenance terminal program 2P.

[0054] The storage unit 21 uses a non-volatile memory such as a hard disk or a flash memory. The storage unit 21 stores various programs including a maintenance terminal program 2P. The storage unit 21 stores screen data based on the maintenance terminal program 2P. The maintenance terminal program 2P may be a program that the control unit 20 reads out from the maintenance terminal program 8P stored in the recording medium 8 and copies to the storage unit 21.

[0055] The first communication unit 22 is a communication device for realizing data communication via the communication network N or the network MN. The first communication unit 22 may be a communication device such as a network card for wired communication. The first communication unit 22 may be a wireless communication device for mobile communication that connects to a base station BS (see FIG. 1). The first communication unit 22 may use a wireless communication device that supports connection to an access point AP.

[0056] The second communication unit 23 is a communication device for establishing data communication by connecting to the maintenance communication device 6. The second communication unit 23 may be a wireless communication device such as Wifi or Bluetooth (registered trademark). The second communication unit 23 may be a Universal Serial Bus (USB) interface.

[0057] The display unit 24 uses a display such as a liquid crystal display or an organic EL (Electro Luminescence) display. The display unit 24 displays an operation screen based on the maintenance terminal program 2P of the control unit 20 and an image of a Web page provided by the maintenance support device 1. The display unit 24 is preferably a display with a built-in touch panel. The display unit 24 may also be a display without a built-in touch panel.

[0058] The operation unit 25 is a keyboard and a pointing device capable of inputting and outputting data to and from the control unit 20. The operation unit 25 may be a user interface such as a voice input unit. The operation unit 25 may be a touch panel of the display unit 24 or physical buttons provided on the housing. The operation unit 25 notifies the control unit 20 of operation information by the user.

[0059] The imaging unit 26 outputs a captured image obtained using an imaging element. The control unit 20 can obtain an image captured by the imaging element of the imaging unit 26 at any timing.

[0060] 3 is a block diagram showing the internal configuration of the maintenance communication device 6. The maintenance communication device 6 includes a control unit 60, a storage unit 61, a first communication unit 62, a second communication unit 63, and a third communication unit 64. The control unit 60 uses a CPU or a microprocessor. The storage unit 61 stores a predefined program.

[0061] A non-volatile memory such as a flash memory is used as the storage unit 61. The storage unit 61 stores the status data received from the energy storage element 50.

[0062] The first communication unit 62 is a communication device that realizes a communication connection with a unit connected to the energy storage element 50. In the present embodiment, the first communication unit 62 communicates with the unit of the energy storage element through wireless communication such as Bluetooth (registered trademark).

[0063] The second communication unit 63 is a communication device that realizes a communication connection via the network CN. The maintenance communication equipment 6 can transmit the status data received from the energy storage element 50 to the management device 51 by the second communication unit 63. If the energy storage element 50 is provided with a battery management device having a communication function, the second communication unit 63 is not necessary.

[0064] The third communication unit 64 is a communication device that realizes a communication connection between the maintenance communication equipment 6 and the maintenance terminal device 2. In the present embodiment, the third communication unit 64 is a USB interface. The third communication unit 64 may be a wireless communication device different from the first communication unit 62.

[0065] The control unit 60 of the maintenance communication device 6 periodically acquires status data from the energy storage element 50 through the first communication unit 62 based on the program. The control unit 60 sequentially stores the acquired status data in the storage unit 61. When the energy storage element 50 is a lead battery, the storage period is, for example, about once a day. The control unit 60 associates the acquired date and time with the status data and stores it in the storage unit 61. The control unit 60 sequentially transmits the acquired status data from the second communication unit 63 to the management device 51. When the control unit 60 is communicatively connected to the maintenance terminal device 2 through the third communication unit 64 based on the program, the control unit 60 reads out the status data from the storage unit 61 in response to an instruction from the maintenance terminal device 2 and transmits the status data from the third communication unit 64.

[0066] The maintenance support system 100 configured as above and the maintenance communication device 6 provided in the power storage device 5 support maintenance management as described below.

[0067] First, status data that is not transmitted to the remote monitoring system 300 via the network CN by the maintenance communication device 6 can also be aggregated in the remote monitoring system 300. The maintenance terminal device 2 carried by a maintenance worker acquires status data stored in the maintenance communication device 6 when regular maintenance inspection is performed. The maintenance terminal device 2 transmits the status data to the remote monitoring system 300 via the network MN or the communication network N, and causes the remote monitoring system 300 to aggregate the status data. Even if a communication connection between the customer's network CN and the remote monitoring system 300 is difficult from a security standpoint, the status data can be aggregated in the remote monitoring system 300 during regular inspection.

[0068] Secondly, the remote monitoring system 300 that has aggregated the status data periodically executes the process described below for each target storage element 50, for example, for each storage device 5 to which multiple storage elements 50 are connected, to determine whether there are any signs of an abnormality (second phase). If the storage element 50 is a lead-acid battery, the execution cycle is three months, six months, etc. The execution cycle is shorter than the periodic inspection cycle. If the storage element 50 is a lithium-ion battery, the execution cycle may similarly be three months, six months, etc. The execution cycle may be shorter depending on the period of use.

[0069] Thirdly, if a sign of an abnormality is detected by the processing of the remote monitoring system 300, the maintenance support device 1 arranges for maintenance work to prevent the trouble before it occurs (third phase). The arranged contents are notified to a maintenance worker. Based on the notified work contents, the maintenance worker goes to the installation site of the power storage device 5 where the sign was detected and carries out the work.

[0070] The processes in the second and third phases will be described in detail below.

[0071] 4 is a flowchart showing an example of a processing procedure for detecting a sign of abnormality in the second phase in the remote monitoring system 300. The remote monitoring system 300 executes the following processing each time a period arrives.

[0072] The remote monitoring system 300 selects some of the multiple storage elements 50 included in the power storage device 5 (step S101). In step S101, the remote monitoring system 300 selects identification data of a storage element group (power storage element group) that corresponds to the identification data of the power storage device 5 to which the power storage element 50 belongs.

[0073] The remote monitoring system 300 acquires state data of the selected energy storage element group (step S102). In step S102, the remote monitoring system 300 acquires, as the state data, for example, the latest data of a voltage value. The state data may be an internal resistance value.

[0074] The remote monitoring system 300 inputs the acquired status data to the judgment model 3M (step S103), and acquires a score related to the judgment output from the judgment model 3M (step S104). The score in this embodiment is a score corresponding to whether or not the input status data includes measurement data of a heterogeneous energy storage element.

[0075] The remote monitoring system 300 determines whether or not the measurement data includes a heterogeneous storage element based on the score output from the judgment model 3M (step S105). If it is determined in step S105 that the measurement data includes a heterogeneous element (S105: YES), the remote monitoring system 300 determines the heterogeneity (step S106). In step S106, the remote monitoring system 300 determines the heterogeneity based on the state data of the multiple storage elements 50 included in the storage device 5 determined to have a warning sign. The judgment model 3M will be described later.

[0076] The remote monitoring system 300 determines whether or not all of the multiple energy storage elements 50 included in the target energy storage device 5 have been selected (step S107). If it is determined that they have not been selected (S107: NO), the remote monitoring system 300 returns the process to step S101.

[0077] If it is determined that all have been selected (S107: YES), the remote monitoring system 300 determines whether or not there is a sign of abnormality in the power storage device 5 based on the heterogeneity determined in step S106 (step S108). The heterogeneity includes things that are not related to abnormality, such as being new or having a long life compared to an assumed model at the time of manufacture. In step S108, the remote monitoring system 300 detects a sign of abnormality when a power storage element 50 with a heterogeneity of having an abnormally short life is included. Since there is a possibility that a long-life power storage element 50 and a new power storage element 50 may cause an unbalanced state with other power storage elements 50 and cause trouble, the remote monitoring system 300 may determine that there is a sign of abnormality for these heterogeneities as well. The determination model 3M may output a score related to the heterogeneity, and the remote monitoring system 300 may execute steps S105 to S108 collectively.

[0078] When it is determined that there is a sign of an abnormality (S108: YES), the remote monitoring system 300 notifies the maintenance support device 1 of a message of the detection of the sign including the identification data of the selected power storage device 5, the identification data of the power storage element group determined to include a heterogeneous power storage element 50, and the heterogeneity (step S109). The remote monitoring system 300 ends the process.

[0079] If it is determined in step S105 that it is not included (S105: NO), and if it is determined in step S108 that there is no sign of abnormality (S108: NO), the remote monitoring system 300 ends the process.

[0080] FIG. 5 shows an overview of the judgment model 3M. In one example, the judgment model 3M uses a convolutional neural network. The judgment model 3M is a classifier that classifies a power storage device 5 including a standard storage element 50 that is not heterogeneous and a power storage device 5 including other heterogeneous storage elements 50. The judgment model 3M includes an input layer 301 that inputs the voltage values ​​of each of the storage elements 50 included in the selected power storage device 5. The judgment model 3M includes an output layer 302 that outputs a score related to the heterogeneity based on the input voltage value. The judgment model 3M includes an intermediate layer 303 that includes a convolutional layer or a pooling layer. The judgment model 3M is trained by providing the neural network with teacher data that includes state data labeled with a label (e.g., "0") that is a standard storage cell that is not heterogeneous and state data labeled with a heterogeneous label (e.g., "1"). The judgment model 3M is trained using the voltage values ​​measured for each of the storage elements 50. The determination model 3M may be trained using the internal resistance value. The determination model 3M outputs a heterogeneity score (a numerical value between 0 and 1) from the output layer 302 for given state data.

[0081] The judgment model 3M is not limited to a classifier, and may be a convolutional neural network that outputs a feature amount. The judgment model 3M may be configured as a network using a recurrent neural network, LSTM (Long Short-term Memory), or the like that inputs time-series data of the measurement data of the same storage element 50 and outputs a feature amount.

[0082] The judgment model 3M may be a model that statistically calculates whether an outlier is included or not and the degree of outlier if an outlier is included, using the average, standard deviation, median, etc. of the voltage value or the internal resistance value. The judgment model 3M may be a model that finds a trend from time-series data of the state data and outputs a score that indicates the degree of heterogeneity according to the difference in the trend. The judgment model 3M may use a k-nearest neighbor algorithm. Using this judgment model 3M, the remote monitoring system 300 may judge whether the target measurement data belongs to a non-heterogeneous class or a heterogeneous class that has been learned in advance based on teacher data. The remote monitoring system 300 may make a judgment by clustering based on a judgment model 3M (judgment program) that uses the k-means method or the EM method. The remote monitoring system 300 may determine whether the target measurement data is heterogeneous by contracting based on a judgment model 3M (judgment program) that uses PCA (Principal Component Analysis).

[0083] The accuracy of the determination as to whether or not a heterogeneous energy storage element 50 is included in the energy storage element group is improved by using the time transition of the output heterogeneity. Fig. 6 is a flowchart showing another example of the processing procedure for detecting a sign of an abnormality in the remote monitoring system 300. Among the processing procedures shown in Fig. 6, steps common to the processing procedures shown in Fig. 4 are assigned the same step numbers and detailed descriptions thereof will be omitted.

[0084] The remote monitoring system 300 stores the degree of heterogeneity acquired in step S104 in association with the identification data and time information of the selected power storage device 5 (step S121).

[0085] The remote monitoring system 300 reads out the stored heterogeneity for the past predetermined period for the power storage device 5 selected in step S101 (step S122). The remote monitoring system 300 creates a time distribution of the heterogeneity for the past predetermined period (step S123).

[0086] The remote monitoring system 300 determines whether or not the energy storage element group includes heterogeneous energy storage cells based on the value of the heterogeneity read out in step S122, the time distribution created in step S123, and / or the state data itself of the energy storage elements 50 included in the selected energy storage device 5 (S105). If it is determined in step S105 that a heterogeneous energy storage cell is included (S105: YES), the remote monitoring system 300 may distinguish the urgency in step S106 using the time distribution created in step S123. The remote monitoring system 300 may determine whether the heterogeneity is one that will cause a problem within one or three months, or one that is expected to continue operation for about six months.

[0087] It is possible to improve detection accuracy by detecting signs of abnormality based on time distribution. There is a possibility that the heterogeneity of the connected energy storage elements will balance out over time and return to normal. Signs of abnormality are detected based on time distribution to prevent this from being mistakenly detected as a sign of abnormality.

[0088] Deep learning may be applied to the process of step S105 itself. The time distribution of heterogeneity may be visualized and input, and whether or not there is heterogeneity and the heterogeneity (type) may be determined based on the pattern of the time distribution. FIG. 7 shows an overview of an image judgment model 32M used to detect signs of anomaly. The image judgment model 32M shown in FIG. 7 receives the time distribution of heterogeneity output from a judgment model 3M that has been trained to output the heterogeneity when state data is input, and outputs a score indicating the accuracy of whether or not a heterogeneous cell is included.

[0089] The image judgment model 32M is a neural network including an intermediate layer including a convolution layer or a pooling layer that extracts features. When an image of a time distribution is input, the image judgment model 32M outputs a probability (score) that includes measurement data of a heterogeneous storage cell in the measurement data related to the time distribution. The image judgment model 32M is stored in the remote monitoring system 300 together with the judgment model 3M. As shown in FIG. 7, the image judgment model 32M is trained by teacher data that is a pair of an image of the time distribution and a result of judgment by an operator.

[0090] When teacher data can be collected, the image determination model 32M may be trained as a model for determining the heterogeneity of heterogeneous energy storage cells. The image determination model 32M may determine, for example, whether the cell is pattern A, pattern B, or pattern C in FIG. 7, that is, the heterogeneity. The remote monitoring system 300 may determine the heterogeneity based on the value of the heterogeneity used in the determination in step S105 in the flowchart of FIG. 6, the time distribution created in step S123, and / or the state data of the selected energy storage device 5. The remote monitoring system 300 may specify the degree of heterogeneity, that is, how far the cell is deviated from the standard energy storage element 50. The remote monitoring system 300 may use the image determination model 32M to determine the heterogeneity, such as whether the cell is a "brand new energy storage cell", a "energy storage element with better quality (longer life) than a standard energy storage element", or a "energy storage element with a shorter life than a standard energy storage element".

[0091] The judgment model 3M has been trained to output the degree of heterogeneity as described above. By targeting heterogeneous energy storage elements 50 that do not match the assumed model at the time of manufacture as replacement targets, it is expected that the energy storage elements 50 included in the energy storage device 5 will match the assumed model at the time of manufacture, thereby improving the accuracy of the deterioration degree or life expectancy prediction. The method of detecting signs of abnormality is not limited to the method using the degree of heterogeneity.

[0092] The determination model 3M may be trained as a model that outputs a discrimination label of the contents of the sign of anomaly and a score indicating the accuracy, without using the degree of heterogeneity. The determination model 3M is trained using accumulated data of the time distribution of state data of a storage element 50 in which an abnormality has been detected in the past and the time distribution of state data of a standard storage element 50 that matches an assumed model at the time of manufacture until the end of its life, as known teacher data.

[0093] FIG. 8 is a flowchart showing an example of a processing procedure in the third phase in the maintenance support system 100. The maintenance support device 1 receives a message of a sign detection from the remote monitoring system 300 (step S201). The control unit 10 determines a necessary maintenance work based on the identification data and heterogeneity of the energy storage element group including the heterogeneous energy storage element 50 contained in the message (step S202). The control unit 10 may determine the necessary maintenance work by referring to the content of the maintenance work associated with the heterogeneity and tabulated in the storage unit 11. The control unit 10 may determine the maintenance work by a learning model that is machine-learned based on the past maintenance work history so as to output the content of the maintenance work (the construction period, the number of workers, and necessary items described later) when the history of heterogeneity or state data is input.

[0094] In step S202, the control unit 10 can appropriately determine the maintenance work contents according to the urgency of the sign when the urgency is distinguished for the heterogeneity. The urgency of the sign is distinguished as whether it is a sign that a trouble will occur within one month or a sign that normal operation will be possible for about six months. It is preferable that the correspondence between the heterogeneity and the maintenance work contents by distinguishing the urgency is stored in the storage unit 11.

[0095] In step S202, the control unit 10 determines the maintenance work to be performed on the energy storage element group including the heterogeneous energy storage element 50 so that the operation of the other energy storage element groups can continue and the energy storage device 5 as a whole can continue to operate. The energy storage device 5 includes multiple energy storage elements 50 and can take measures to continue operation as a whole even if an abnormality occurs in one part.

[0096] The control unit 10 determines the period of the determined maintenance work, the number of workers, and items including replacement parts or tools required for the maintenance work (step S203). In step S203, the control unit 10 may determine the period of the determined maintenance work, based on a table as described above, or may determine the items based on machine learning.

[0097] The control unit 10 creates quotation data based on the determined construction period, number of workers, and necessary items (step S204). The storage unit 11 stores the construction period, unit price for workers, amount of work, and cost of necessary items for creating the quotation data. The control unit 10 refers to this and automatically creates the quotation data. The unit price may be changed depending on the skill of the worker.

[0098] The control unit 10 notifies the sales representative of the target power storage device 5 of the created quotation data or an implementation approval request including a link to the data (step S205).

[0099] The maintenance terminal device 2 used by the sales representative receives a request for approval to perform maintenance work (step S301). The sales representative submits an estimate to the customer based on the estimate data included in the received request for approval to perform maintenance work, and receives approval to perform the maintenance work. If approval is received, the control unit 20 of the maintenance terminal device 2 accepts an approval operation in accordance with the operation of the operation unit 25 (step S302). The control unit 20 accepts, together with the approval operation, an input of a schedule candidate based on agreement with the customer via the operation unit 25 (step S303). The control unit 20 transmits the accepted approval to perform maintenance and the schedule candidate to the maintenance support device 1 (step S304).

[0100] The maintenance support device 1 receives the implementation approval and the schedule candidate from the maintenance terminal device 2 (step S206). The control unit 10 determines the worker and implementation date and time of the maintenance work based on the schedule received in step S206, the operation information of the power storage device 5 based on the customer data management system 400, the list of maintenance workers, and the schedule data of each worker (step S207). In step S207, the control unit 10 may determine the implementation date and time according to the urgency. In step S207, the control unit 10 can determine the worker and implementation date and time based on the operation information and the schedule approved by the customer so that the operation of the power storage device 5 as a whole can be continued using the other power storage element groups excluding the power storage element group including the heterogeneous power storage element 50.

[0101] The control unit 10 notifies the determined worker of the determined implementation date and time, necessary items, and work content (step S208). The control unit 10 orders the necessary items (step S209), notifies the order information to the worker (step S210), and ends the process. The order information in step S209 may be registered as customer assets in the customer data management system 400.

[0102] In this way, at the stage of signs of abnormalities that are difficult for humans to detect, the details of maintenance work to prevent problems are decided and even estimates are prepared, reducing the burden on maintenance workers.

[0103] The embodiments disclosed above are illustrative in all respects and are not restrictive. The scope of the present invention is defined by the claims, and includes all modifications within the meaning and scope of the claims. [Explanation of symbols]

[0104] 100 Maintenance Support System 1 Maintenance support equipment 10 Control section 11 Storage section 1P Maintenance Support Program 2 Maintenance terminal equipment 20 Control section 21 Memory section 2P Maintenance terminal program 300 Remote Monitoring System 3M Judgment Model 32M image judgment model 400 Customer Data Management System 6 Maintenance communication equipment

Claims

1. Detecting a sign of an abnormality in the storage element based on measurement data related to the storage element sequentially stored in the storage device; determining at least one of a duration of a maintenance work corresponding to the detected abnormality sign, the number of workers, and items including replacement items or tools required for the maintenance work; Notify the maintenance worker of the decision and the implementation of the corresponding maintenance work. Maintenance support method.

2. The storage element is configured to include a plurality of storage elements, The detection of the abnormality sign is performed for each of the plurality of storage elements or for each group of the plurality of storage elements, For a storage element or group of storage elements where a warning sign has been detected, the work period, number of workers or items, and the workers and work date and time are determined so that the operation of other storage elements can continue. The maintenance support method according to claim 1 .

3. using a judgment model that has been trained to output a score corresponding to whether or not the measurement data includes measurement data of a heterogeneous storage element when measurement data measured for each storage element or for each storage element group including a plurality of the storage elements is input; When it is determined that the measurement data includes a heterogeneous energy storage element based on the score output by the determination model, the heterogeneity is determined; A sign of an abnormality in the storage element is detected based on the determined heterogeneity. The maintenance support method according to claim 1 or 2.

4. creating a time distribution of the scores output from the determination model in response to input of the measurement data of the energy storage element group; Differentiate urgency and identify heterogeneity based on the time distribution created The maintenance support method according to claim 3.

5. using an image determination model that has been trained to output a score corresponding to whether or not measurement data of a storage element or a storage element group corresponding to the time distribution includes measurement data of a heterogeneous storage element when an image of the created time distribution is input; The created time distribution is visualized and input to the image judgment model; The heterogeneity of the energy storage element group is determined based on the score output from the image determination model. The maintenance support method according to claim 4.

6. determining a worker and a date and time for carrying out the maintenance work based on a list of workers who can carry out the maintenance work and schedule data for each of the workers; Notifying the determined worker of the execution of the maintenance work. The maintenance support method according to any one of claims 1 to 5.

7. receiving approval for carrying out the maintenance work from an owner of the energy storage element; If the approval is received, notify the maintenance work to be performed. The maintenance support method according to any one of claims 1 to 6.

8. If the implementation approval is received, a date and time for implementation is determined based on the operation information of the device obtained from the owner. The maintenance support method according to claim 7.

9. When the implementation approval is received, an order is placed for items necessary for the maintenance work from a vendor of the items; Notifying the worker of the identification data of the ordered item The maintenance support method according to claim 7 or 8.

10. Create estimate data for the maintenance work based on the determined construction period, number of workers, and necessary items; Accepting the implementation approval based on the quotation data The maintenance support method according to any one of claims 7 to 9.

11. A storage device that periodically acquires and sequentially stores measurement data relating to the energy storage element, a maintenance terminal device that can be connected to the storage device, and a maintenance support device that can be connected to the maintenance terminal device for communication, The maintenance support device includes: When an abnormality or a sign of an abnormality is detected in the energy storage element based on the measurement data related to the energy storage element, determining at least one of a construction period, a number of workers, and an item including a replacement item or a tool required for the maintenance work related to the detected abnormality or sign of an abnormality; Send instructions for carrying out the maintenance work, including the decision, to the maintenance worker. Maintenance support system.

12. a determination unit that, when a sign of an abnormality in the energy storage element is detected, determines at least one of a period of maintenance work, a number of workers, and an item including a replacement part or a tool required for the maintenance work, based on measurement data related to the energy storage element that is sequentially stored in a storage device in association with identification data for identifying the energy storage element; A transmission unit that transmits an implementation instruction including a decision item to a worker of the maintenance work; A maintenance support device comprising:

13. On the computer, Detecting a sign of an abnormality in the energy storage element based on measurement data relating to the energy storage element sequentially stored in a storage device in association with identification data for identifying the energy storage element; determining at least one of a duration of a maintenance work corresponding to the detected abnormality sign, the number of workers, and items including replacement items or tools required for the maintenance work; Notify the maintenance worker of the decision and the implementation of the corresponding maintenance work. A computer program that executes a process.

Images