A power supply service AI assistant system and method based on whole-process time limit early warning

Abstract

The application belongs to the technical field of power system automation and artificial intelligence application, and is a power supply service AI assistant system based on whole-process time limit early warning, characterized by comprising: a work order access module, which is used for receiving fault work orders from a 95598 customer service system and an active repair system; a work order execution module, which automatically distributes work orders to the terminals of repair personnel in corresponding grids and records the time of distributing work orders; a time limit rule library, which is used for storing time limit thresholds of each link; a whole-process early warning module, which is used for monitoring the flow state of work orders in each link; and a data feedback module, which is used for recording time limit execution data of each work order in the whole process, generating an overtime risk analysis report, and providing data support for parameter optimization of the time limit rule library. The problems of insufficient whole-process time limit control of fault repair work orders, single follow-up mechanism and process overtime of repair personnel caused by on-site factors in the prior art can be solved.

Description

Technical Field

[0001] This invention belongs to the field of power system automation and artificial intelligence application technology, specifically relating to an AI virtual assistant system for power supply service command and its usage method, used to realize the automatic dispatch of fault repair work orders and the full-process time limit graded early warning. Background Technology

[0002] With the deepening of the digital transformation of the power system, power supply companies across the country are exploring the application of AI technology to improve the efficiency of power supply service command. Among existing technologies, some power supply companies have already deployed work order monitoring systems or AI virtual call center assistants based on RPA (Robotic Process Automation). For example, the State Grid Qinghai Electric Power Company's publicly disclosed patent (CN119444239A) for an "AI Virtual Call Center Assistant for Power Supply Services" proposes modules such as event monitoring, intelligent duty, and auxiliary command, enabling automatic work order dispatch and automatic dispatch of SMS reminders for overdue work orders. Another example is the "Dian Xiao Fu" intelligent call center assistant developed by the Fuzhou Power Supply Company, which can automatically dispatch work orders and push repair progress updates. The Taining Power Supply Company has also implemented a full-chain mechanism of "intelligent power outage assessment—automatic work order generation—precise grid-based push," reducing dispatch time to less than 3 minutes.

[0003] However, the aforementioned existing technologies still have the following shortcomings in practical applications: First, the existing system only provides a one-time reminder function, that is, a reminder is sent when the work order is about to expire or has already expired, lacking a tiered early warning mechanism for the entire work order process. Due to the complexity of the situation at the repair site, front-line repair personnel are often too busy with on-site operations or poor signal to complete the process confirmation of order acceptance, arrival, and repair in the system in a timely manner, resulting in work order process delays. This not only affects service performance indicators, but also puts a significant psychological burden on repair personnel.

[0004] Secondly, existing technologies primarily assist on-duty personnel in the automatic work order dispatching process, but lack proactive and intelligent monitoring and intervention throughout the entire process after dispatching. On-duty personnel still need to manually monitor the timelines and progress of multiple work orders, which can easily lead to oversights during periods of high failure rates.

[0005] Third, existing reminder methods are mostly one-way notifications, lacking awareness and interactive confirmation of the status of emergency repair personnel, and cannot effectively determine whether the reminder information has been received or whether the emergency repair personnel are experiencing actual difficulties. Summary of the Invention

[0006] The purpose of this invention is to provide a power supply service AI assistant system and method based on full-process time limit early warning, so as to solve the problems of insufficient full-process time limit control of fault repair work orders, single reminder mechanism, and easy time for repair personnel to exceed the process time due to on-site factors in the existing technology.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a power supply service AI assistant system based on full-process time-limit early warning, characterized in that it includes: The work order access module is used to receive fault work orders from the 95598 customer service system and the proactive repair system. The work order includes the fault address, fault type, user information and work order generation time. The dispatch execution module, connected to the work order access module, is used to automatically dispatch work orders to the terminals of emergency repair personnel in the corresponding grid according to the preset grid dispatch rules, and record the dispatch time. The time limit rule library is used to store the time limit thresholds for each stage, including the order acceptance time limit, the on-site arrival time limit (distinguishing between urban and rural areas), and the fault repair time limit (distinguishing between urban and rural areas and remote areas). The input end of the time limit rule library is connected to the terminal communication of the emergency repair personnel and the reverse confirmation mechanism submodule of the full-process early warning module, respectively. The output end of the time limit rule library is connected to the data feedback module, respectively. The end-to-end early warning module is connected to both the work order execution module and the time limit rule base. It monitors the workflow status of work orders at each stage and executes tiered early warnings based on the ratio of the time elapsed at the current stage to the corresponding time limit threshold. Specifically, the end-to-end early warning module includes: 1) Order Acceptance Early Warning Submodule: Starting from the order dispatch time, when the time consumed in the order acceptance process reaches the first preset percentage of the order acceptance time limit, a first-level order acceptance reminder is sent to the emergency repair personnel's terminal; when the second preset percentage of the order acceptance time limit is reached, a second-level order acceptance reminder is sent to both the emergency repair personnel's terminal and the on-duty personnel's terminal; when the order acceptance process times out, an overtime record is generated and pushed to the on-duty personnel's terminal. 2) On-site arrival warning submodule: Starting from the order confirmation time, the module retrieves the corresponding arrival time limit based on the work order's region type (urban or rural) and sends it to the on-site warning system according to a preset ratio. 3) Repair process early warning submodule: Starts timing from the arrival confirmation time, retrieves the corresponding fault repair time limit according to the area type of the work order, and sends repair early warnings according to preset proportions and levels; 4) Multi-channel reach module, which is connected to the order receiving stage warning sub-module, the on-site stage warning sub-module, and the repair stage warning sub-module of the full-process warning module, and is connected to the emergency repair personnel terminal. It is used to send warning information simultaneously through three methods: SMS, mobile application push (APP push), and voice outbound call (interacting with the emergency repair personnel terminal), and to receive the confirmation receipt from the emergency repair personnel terminal. 5) The reverse confirmation mechanism submodule is used to automatically upgrade to a telephone outbound call and transfer to the on-duty personnel if no confirmation receipt is received from the repair personnel's terminal within a preset time after sending the warning information, so that the on-duty personnel can confirm manually. The data feedback module records the time limit execution data of each work order throughout the entire process, generates a timeout risk analysis report, and provides data support for parameter optimization of the time limit rule base.

[0008] Furthermore, in the full-process early warning module, the preset ratio of the graded early warning can be configured as follows: the first preset ratio is 50%, and the second preset ratio is 80%.

[0009] Furthermore, in the reverse confirmation mechanism submodule, the preset confirmation time is 3 minutes.

[0010] Furthermore, the system also includes an intelligent judgment assistance module, which automatically retrieves the current geographical location of the repair personnel associated with the work order and historical work order processing efficiency data when an early warning is triggered, to assist the on-duty personnel in determining whether additional support or adjustments to repair resources are needed; the intelligent judgment assistance module is connected to the time limit rule library and the full-process early warning module respectively.

[0011] The above-mentioned method for using a power supply service AI assistant system based on full-process time limit early warning is characterized by the following steps: 1) Prepare a power supply service AI assistant system based on full-process time limit early warning, as described above; 2) Obtain the current status and elapsed time of the work order; 3) Retrieve the corresponding time limit threshold from the time limit rule library; 4) Calculate the ratio of the elapsed time to the time limit threshold; ① When the ratio is ≥50% and <80%, trigger the first-level warning; ② When the ratio is ≥80% and <100%, trigger the second-level warning; ③ When the ratio is ≥100%, determine that the time has expired, generate an overtime record and push it to the terminal of the on-duty personnel. 5) Has the work order been transferred to the next stage? ① Yes, reset the timer; ② No, end.

[0012] Compared with the prior art, the present invention has the following beneficial effects: 1. Full-process hierarchical early warning to avoid timeouts at single nodes: This invention extends work order management from "order completion" to the entire process of "order acceptance - arrival - repair". A hierarchical early warning mechanism is set up for each link to proactively remind the personnel before the time limit is reached, giving the repair personnel sufficient response time and effectively avoiding timeout problems caused by neglecting system process operations due to being busy with on-site operations.

[0013] 2. Multi-channel reach and reverse confirmation ensure information delivery: This invention simultaneously sends alerts via SMS, APP push notifications, and voice calls, and ensures effective reception of alert information through a reverse confirmation mechanism. If repair personnel fail to confirm in a timely manner, the system automatically escalates to manual intervention, avoiding communication failures caused by signal or equipment problems with a single communication method.

[0014] 3. Reduce the dual burden on on-duty personnel and emergency repair personnel: On-duty personnel do not need to manually monitor the time-limit progress of a large number of work orders, as the system automatically completes the entire process monitoring and early warning; Emergency repair personnel can focus on on-site emergency repairs while completing system process operations in a timely manner through early warning reminders, avoiding the assessment pressure and psychological burden caused by process timeouts.

[0015] 4. Data-driven rule optimization: By recording the execution data of the entire process time limit, the system can automatically analyze the high-incidence time periods, high-incidence areas, and high-incidence personnel of timeouts in each link, providing data support for the dynamic adjustment of the time limit rule base and the optimized allocation of emergency repair resources, forming a closed loop of continuous improvement. Attached Figure Description

[0016] Figure 1 This is a structural block diagram of the system of the present invention.

[0017] Figure 2 This is a flowchart of the full-process early warning module of the present invention.

[0018] Figure 3 This is a time-series diagram illustrating the tiered early warning system of the present invention. Detailed Implementation

[0019] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0020] Example 1 This embodiment provides a power supply service AI assistant system based on full-process time-limit early warning, such as... Figure 1 As shown, it includes a work order access module, a work order execution module, a time limit rule library, a full-process early warning module, and a data feedback module.

[0021] Work order access module: Receives fault repair work orders from the 95598 customer service system and proactive repair work orders from the power distribution automation system in real time. Taking a specific fault as an example, the system received a fault work order for the urban area at 10:00:00, containing key information such as the fault address and the user's contact information.

[0022] Work order execution module: Based on the preset grid-based work order dispatch rules, it automatically identifies the emergency repair grid to which the fault address belongs, dispatches the work order to the mobile terminal of Zhang, the emergency repair personnel of that grid, at 10:00:30, and records the dispatch time as 10:00:30.

[0023] Time limit rules: The preset order acceptance time limit is 5 minutes, the on-site arrival time limit in urban areas is 45 minutes, and the fault repair time limit in urban areas is 3 hours.

[0024] Full-process early warning module: Starts timing from 10:00:30 when the order is dispatched.

[0025] Order acceptance warning: When the timer reaches 10:02:30 (50% of the 5-minute order acceptance time limit), the order acceptance warning submodule triggers the first-level order acceptance reminder; if no order is accepted by 10:04:30 (80% of the order acceptance time limit), the second-level order acceptance reminder is triggered; if no order is accepted by 10:05:30, the order acceptance is judged to have timed out, an overtime record is generated and pushed to the duty personnel's terminal.

[0026] After the order was confirmed, the repairman, Zhang, clicked "Accept Order" on the terminal at 10:05:00, and the system recorded the order acceptance time. The on-site arrival warning submodule started timing at 10:05:00. When the timer reached 10:27:30 (50% of the 45-minute arrival time limit for urban areas), the first-level on-site arrival warning was triggered; at 10:41:00 (80%), the second-level on-site arrival warning was triggered.

[0027] Repairman Zhang arrived at the scene at 10:40:00 and clicked "Arrival Confirmation," and the system recorded the arrival time. The repair process early warning submodule started timing at 10:40:00 and sent repair warnings in a tiered manner according to a preset ratio.

[0028] Multi-channel reach module and reverse confirmation mechanism: Warning information is simultaneously sent to the mobile phones of emergency repair personnel via SMS, pushed to the emergency repair terminal via mobile application, and broadcast via voice outbound call. If no confirmation is received from the emergency repair personnel on the terminal within 3 minutes after the warning is sent, the reverse confirmation mechanism submodule automatically escalates the warning to a telephone outbound call and transfers the call to the on-duty personnel, who then manually confirm the status of the emergency repair personnel.

[0029] Example 2 The difference between this embodiment and Embodiment 1 is that it further includes an intelligent judgment assistance module. When a work order triggers a second-level warning during the on-site arrival phase, the intelligent judgment assistance module automatically retrieves the current geographical location of the repair personnel to determine whether they are near the fault point; simultaneously, it retrieves the personnel's historical work order processing efficiency data to assess their ability to arrive on time. If a risk of timeout is determined, the system automatically pushes "suggest dispatching additional support" auxiliary decision-making information to the on-duty personnel, who can then allocate surrounding repair resources according to the actual situation.

[0030] Example 3 In this embodiment, the threshold of the time limit rule base can be configured differently according to different regions and time periods. For example, for key power supply areas during peak tourist seasons, the on-site arrival time can be adjusted to 30 minutes; for remote mountainous areas, the repair time can be appropriately extended. The preset proportion of tiered early warning can also be configured according to actual needs, and is not limited to 50% and 80%.

[0031] Industrial Applicability: This invention can be widely applied in power supply service command centers of power supply companies at all levels, serving as an intelligent upgrade module for existing power supply service command systems. The system can be deployed in a lightweight manner, and can interface with existing marketing systems, PMS systems, and distribution automation systems, resulting in low implementation costs and high promotional value.

Claims

1. A power supply service AI assistant system based on full-process time-limit early warning, characterized in that... include: The work order access module is used to receive fault work orders from the 95598 customer service system and the proactive repair system. The work order includes the fault address, fault type, user information and work order generation time. The dispatch execution module, connected to the work order access module, is used to automatically dispatch work orders to the terminals of emergency repair personnel in the corresponding grid according to the preset grid dispatch rules, and record the dispatch time. The time limit rule library is used to store the time limit thresholds for each stage, including the order acceptance time limit, the arrival time limit on site, and the fault repair time limit. The input end of the time limit rule library is connected to the terminal communication of the emergency repair personnel and the reverse confirmation mechanism submodule of the full-process early warning module, respectively. The output end of the time limit rule library is connected to the data feedback module. The full-process early warning module is connected to the dispatch execution module and the time limit rule base respectively. It is used to monitor the flow status of work orders at each stage and to execute graded early warnings based on the ratio of the time consumed in the current stage to the corresponding time limit threshold. The full-process early warning module specifically includes: 1) Order Acceptance Early Warning Submodule: Starting from the order dispatch time, when the time consumed in the order acceptance process reaches the first preset percentage of the order acceptance time limit, a first-level order acceptance reminder is sent to the emergency repair personnel's terminal; when the second preset percentage of the order acceptance time limit is reached, a second-level order acceptance reminder is sent to both the emergency repair personnel's terminal and the on-duty personnel's terminal; when the order acceptance process times out, an overtime record is generated and pushed to the on-duty personnel's terminal. 2) On-site arrival warning submodule: Starts counting from the order confirmation time, retrieves the corresponding arrival time limit according to the area type of the work order, and sends on-site warnings according to the preset proportions and levels; 3) Repair process early warning submodule: Starts timing from the arrival confirmation time, retrieves the corresponding fault repair time limit according to the area type of the work order, and sends repair early warnings according to preset proportions and levels; 4) Multi-channel reach module, which is connected to the order receiving stage warning sub-module, the on-site stage warning sub-module, and the repair stage warning sub-module of the full-process warning module, and is connected to the terminal of the emergency repair personnel. It is used to send warning information simultaneously through three methods: SMS, mobile application push and voice outbound call, and to receive the confirmation receipt from the terminal of the emergency repair personnel. 5) The reverse confirmation mechanism submodule is used to automatically upgrade to a telephone outbound call and transfer to the on-duty personnel if no confirmation receipt is received from the repair personnel's terminal within a preset time after sending the warning information, so that the on-duty personnel can confirm manually. The data feedback module records the time limit execution data of each work order throughout the entire process, generates a timeout risk analysis report, and provides data support for parameter optimization of the time limit rule base.

2. The power supply service AI assistant system based on full-process time-limit early warning as described in claim 1, characterized in that... In the full-process early warning module, the preset ratio of the graded early warning can be configured as follows: the first preset ratio is 50%, and the second preset ratio is 80%.

3. The power supply service AI assistant system based on full-process time-limit early warning as described in claim 1, characterized in that... In the reverse confirmation mechanism submodule, the preset confirmation time is 3 minutes.

4. The power supply service AI assistant system based on full-process time-limit early warning as described in claim 1, characterized in that... The system also includes an intelligent analysis and assistance module, which automatically retrieves the current geographical location of the repair personnel associated with the work order and historical work order processing efficiency data when an early warning is triggered, to assist the on-duty personnel in determining whether additional support or repair resources need to be dispatched. The intelligent analysis and assistance module is connected to the time limit rule library and the full-process early warning module.

5. The method of using the power supply service AI assistant system based on full-process time-limit early warning as described in claim 1, characterized in that... Includes the following steps: 1) Prepare a power supply service AI assistant system based on full-process time limit early warning; 2) Obtain the current status and elapsed time of the work order; 3) Retrieve the corresponding time limit threshold from the time limit rule library; 4) Calculate the ratio of the elapsed time to the time limit threshold; ① When the ratio is ≥50% and <80%, trigger the first-level warning; ② When the ratio is ≥80% and <100%, trigger the second-level warning; ③ When the ratio is ≥100%, determine that the time has expired, generate an overtime record and push it to the terminal of the on-duty personnel. 5) Has the work order been transferred to the next stage? ① Yes, reset the timer; ② No, end.

Images