Short circuit identification method and device for terminal electricity utilization acquisition
By collecting current waveform characteristics and performing multi-cycle comparisons, the accuracy and reliability issues of short-circuit identification in end-user power acquisition equipment have been resolved, achieving efficient short-circuit identification and location without a mechanical tripping structure.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ACREL CO LTD
- Filing Date
- 2025-12-09
- Publication Date
- 2026-05-12
AI Technical Summary
Existing technologies cannot accurately identify short circuits in end-point power consumption data acquisition devices without relying on mechanical tripping structures, and the current threshold-based method has a low recognition rate and cannot capture the complete short-circuit current waveform.
By acquiring current signals in real time, constructing a multi-level buffer queue, extracting peak-shaving and shoulder-flattening features of the current waveform, and performing multi-cycle joint comparison with a preset short-circuit feature rule base, combined with state consistency judgment to achieve short-circuit identification.
It achieves accurate and reliable identification of short circuits without a mechanical tripping structure, improves the timeliness and anti-interference ability of identification, provides short circuit location information, and improves operation and maintenance efficiency and identification accuracy.
Smart Images
Figure CN122017666A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to power system fault detection technology, and in particular to a short-circuit identification method and device for end-point power consumption data acquisition. Background Technology
[0002] Currently, there are two main short-circuit identification solutions for end-point power consumption data acquisition equipment on the market. One method utilizes the trip unit and bimetallic strip inside the smart circuit breaker. The trip unit uses a high-current coil triggering mechanism to trip, while the bimetallic strip uses a high current to heat up the bimetallic strip, causing the mechanical mechanism to trip and achieving short-circuit protection. This method can achieve rapid short-circuit protection, but it requires the mechanical structure of the load to support it. If only the data acquisition component is used, short-circuit protection cannot be achieved. The second method uses the acquired current to set a specific current threshold. By setting a relatively high current, the magnitude of the current is roughly judged to trigger an alarm. This method is only suitable for certain specific scenarios. In many cases, the short-circuit current may not reach the preset threshold, resulting in a low recognition rate. Furthermore, when a regular circuit breaker trips quickly, the data acquisition device often cannot capture the complete short-circuit current waveform in time, thus failing to effectively identify the short-circuit event. This makes it difficult for maintenance personnel to quickly locate the cause of the trip, and still requires tedious manual inspection of the downstream load and lines.
[0003] A search revealed Chinese Patent Publication No. CN119667364A, which discloses a method for rapid identification of short-circuit faults in feeder terminal equipment. This method involves collecting current signals and setting two threshold values: an instantaneous current value setting value and a current change rate setting value. A fault database is constructed, and dual-threshold logic is used for judgment to achieve rapid identification of short-circuit faults. However, this method is essentially still a threshold judgment method based on current amplitude and change rate, and the feature quantities used are relatively limited, failing to fully utilize the morphological characteristics of the short-circuit waveform.
[0004] Therefore, how to accurately identify short circuits at the end without relying on mechanical tripping structures, solely by analyzing current waveform characteristics, is a technical problem that needs to be solved. Summary of the Invention
[0005] The purpose of this invention is to overcome the defects of the prior art and provide a short-circuit identification method and device for end-point power consumption data acquisition.
[0006] The objective of this invention can be achieved through the following technical solutions: According to a first aspect of the present invention, a short-circuit identification method for end-point power consumption acquisition is provided, comprising: The current signal in the power supply circuit is acquired in real time and converted into continuous cycle current waveform data; The current waveform data of a continuously preset number of cycles are cached in parallel to form a multi-level cache queue; Waveform features of a consecutive preset number of cycles are synchronously extracted from the multi-level buffer queue; the waveform features include peak clipping features describing current limiting characteristics and shoulder flattening features describing current steady-state characteristics; The waveform features are compared with a preset short-circuit feature rule base over multiple cycles. Based on the consistency of the multi-cycle joint comparison results, it is determined whether a short circuit has occurred.
[0007] As a preferred technical solution, the parallel caching of current waveform data for a predetermined number of consecutive cycles specifically includes: Set a preset number of FIFO buffers to be used cyclically; The latest cycle data acquired in real time is stored in the current buffer, and the data in each buffer of the previous cycle is moved to the next cycle in sequence to form the multi-level cache queue.
[0008] As a preferred technical solution, extracting the waveform features specifically includes: Identify the amplitude-limiting segments in the current waveform that exceed a preset current threshold, count them as the number of peak-clipping waveforms, and count the number of sampling points for the duration of the amplitude-limiting segments to obtain the peak-clipping duration period; Identify the steady-state segments in the current waveform that are within a preset current range, count them as the number of flat-shoulder waves, and count the number of sampling points that the steady-state segments last for a period of time to obtain the flat-shoulder wave period.
[0009] As a preferred technical solution, before extracting the waveform features, the method further includes: Calculate the similarity between the current cycle current waveform data and the pre-stored typical short-circuit waveform; If the similarity is lower than a preset threshold, the current waveform is determined to be a normal load waveform, and the subsequent process is terminated. Otherwise, the step of extracting waveform features is triggered.
[0010] As a preferred technical solution, the process of determining whether a short circuit has occurred specifically includes: If the waveform characteristics of the current cycle match the preset short-circuit feature rule base, a pre-alarm status is generated. The pre-alarm status generated by subsequent cycles is compared with the pre-alarm status of previous cycles for consistency. If the alarm status for a consecutive preset number of cycles indicates a short circuit, then a short circuit is confirmed to have occurred.
[0011] As a preferred technical solution, establishing the preset short-circuit feature rule base specifically includes: Short-circuit experiments were conducted at different short-circuit distances, and current waveform data were collected when the short circuit occurred. Statistical analysis of the current waveform data is used to establish a model of the correspondence between short-circuit distance and waveform characteristic quantities.
[0012] As a preferred technical solution, after confirming that a short circuit has occurred, the method further includes: Based on the identified flat shoulder wave, its steady-state current value is obtained, and the steady-state current value is determined as the alarm current value for this short circuit event.
[0013] As a preferred technical solution, the method further includes: After each short circuit is determined and confirmed on-site, record the waveform characteristics of this short circuit event and the corresponding actual short circuit distance. Based on the newly recorded data, the distance intervals and feature value ranges in the preset short-circuit feature rule base are dynamically corrected.
[0014] According to a second aspect of the present invention, an apparatus is provided for implementing the short-circuit identification method for end-point power consumption acquisition, the apparatus comprising: The data acquisition module is used to acquire the current signal in the power supply circuit in real time and convert it into current waveform data; The feature extraction module is used to extract waveform features from the current waveform data. The waveform features include peak clipping features for describing current limiting characteristics and shoulder flattening features for describing current steady-state characteristics. The processing and judgment module is used to compare the extracted waveform features with a preset short-circuit feature rule library, and determine whether a short circuit has occurred based on the comparison result.
[0015] As a preferred technical solution, the device further includes: The caching module uses multiple buffers to cache the current waveform data, forming a multi-level cache queue; The communication module is used to send the generated alarm information to the gateway via wireless communication after the processing and judgment module determines that a short circuit has occurred. The alarm information includes the alarm current value.
[0016] Compared with the prior art, the present invention has the following advantages: 1. This invention analyzes the multi-dimensional morphological characteristics of current waveforms, such as peak clipping and flat shoulders, and performs short circuit identification based on the joint comparison of multi-cycle waveform characteristics. It can achieve accurate and reliable identification of terminal short circuits without the need for a mechanical tripping structure.
[0017] 2. This invention processes continuous cycle data by setting up multi-level buffer queues, thereby achieving real-time and continuous tracking and analysis of short-circuit waveform characteristics, ensuring the integrity of fault feature capture and the timeliness of identification.
[0018] 3. The present invention adopts a multi-cycle joint comparison and state consistency judgment mechanism, which effectively distinguishes between real continuous short circuits and transient interference, and improves the anti-interference ability of the system and the reliability of the identification results.
[0019] 4. This invention can further estimate the distance of the short circuit point based on the matching relationship between waveform characteristics and preset rules, providing key information for rapid fault location and troubleshooting, and improving operation and maintenance efficiency.
[0020] 5. This invention can continuously improve the feature rule base by accumulating field data, thereby enhancing adaptability and recognition accuracy under different lines and environments. Attached Figure Description
[0021] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a schematic diagram of the parallel buffering of current waveform data in this invention; Figure 3 This is a schematic diagram of the multi-cycle joint comparison and judgment of the present invention; Figure 4 This is a system architecture diagram of the present invention; Detailed Implementation
[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.
[0023] Example 1: like Figure 1 As shown, this invention provides a short-circuit identification method for end-point power consumption data acquisition, the method comprising the following steps: Step S1: Data Acquisition and Caching The high-precision analog-to-digital converter (ADC) integrated inside the microcontroller (MCU) is used to acquire the AC current signal in the power supply circuit in real time and convert it into digital current waveform data. Approximately 64 data points are acquired for each AC cycle. The current waveform data of four consecutive cycles is cached in parallel, specifically through four cyclically used buffers buf1-buf4, forming a multi-level cache queue. The caching strategy is as follows: Figure 2 As shown, specifically: the latest acquired cycle data is stored in buf1, the original buf1 data is moved to buf2, the original buf2 data is moved to buf3, and so on, to ensure that buf3-buf6 always contain the latest waveform data of 4 consecutive cycles.
[0024] Step S2: Waveform Feature Extraction To improve processing efficiency, before proceeding to feature extraction, the latest cycle data in buf1 can be quickly filtered, and the cosine similarity between the cycle data and the typical short-circuit waveform template pre-stored in ROM can be calculated. If the similarity is lower than the preset threshold, the current waveform is considered to be a normal load start-up or running waveform, and the subsequent process of this cycle is terminated, waiting for the next cycle. If the similarity is higher than the threshold, the subsequent fine feature extraction steps are triggered. Waveform features of these four consecutive cycles are extracted synchronously from Buf3 to Buf6 of the multi-level buffer queue. The features include: Peak-shaving characteristics: Peak clipping waveform count: The number of waveforms in the current cycle whose peak value is limited to a certain level. Peak shaving duration: The number of sampling points where peak shaving continues; Flat shoulder characteristics: Flat-shoulder wave count: The number of waveforms in the current cycle where the current remains at a relatively high steady-state value; Shoulder wave period: Count the number of sampling points for each shoulder wave duration.
[0025] Step S3: Multi-period joint comparison and judgment The extracted waveform features of four consecutive cycles were compared with a pre-defined short-circuit feature rule base. This rule base was established through extensive prior experiments, and a correspondence model between short-circuit distance and waveform features was established after statistical analysis. Some data are shown in Table 1. Table 1 The comparison and judgment process is as follows: Figure 3 As shown, specifically: If the feature value of the current cycle matches a rule in the rule base, a pre-alarm status is generated for that cycle. The pre-alarm status generated by subsequent cycles is compared with the pre-alarm status of previous cycles for consistency. If the alarm status for a consecutive preset number of cycles indicates a short circuit, then the short circuit is finally confirmed to have occurred.
[0026] Step S4: Self-learning optimization After confirming the short circuit and having the on-site maintenance personnel verify the actual short circuit location, the system executes a self-learning optimization process: Record the complete waveform characteristics of this short circuit event, including the number of clipped peaks, the duration of clipping peaks, the number of flat-shoulder waves, the duration of flat-shoulder waves, and the actual short circuit distance confirmed on site; The newly recorded data points are combined with the data in the original rule base for regression analysis, and the range of feature values for the corresponding distance intervals in the rule base is dynamically adjusted. The updated rule base is stored in non-volatile memory for subsequent short-circuit identification.
[0027] Step S5: Alarm Reporting After confirming that a short circuit has occurred, perform the following operations: Alarm reporting: Generate alarm information, including the alarm event, alarm time, and alarm current value obtained from the flat shoulder wave, and broadcast the alarm information to the home smart gateway through the 2.4GHz wireless communication module.
[0028] The method of this invention acquires current waveforms in real time and constructs a multi-level buffer queue. It simultaneously extracts peak clipping and flat-shoulder waveform features from multiple consecutive cycles, and then performs multi-cycle joint comparison with a preset rule base. Finally, it accurately judges short circuits based on the consistency of the comparison results, thereby achieving high-precision, low-false-alarm end-circuit identification.
[0029] Example 2: like Figure 4 As shown, this invention provides an apparatus for implementing a short-circuit identification method for end-point power consumption data acquisition. This apparatus is integrated within an end-point power consumption data acquisition device and mainly includes: Data acquisition module 1: Real-time acquisition of current signals in the power supply circuit and conversion of them into digital current waveform data; Cache module 2: Implemented by multiple FIFO buffers partitioned from the MCU's internal RAM, used to form a multi-level cache queue to store continuous current waveform data; Feature extraction module 3: Synchronously extracts peak clipping and shoulder flattening features of multiple consecutive cycles from the multi-level cache queue of cache module 2; Processing and Judgment Module 4: It has a built-in short-circuit feature rule library, performs multi-cycle joint comparison and consistency judgment of waveform feature quantities, and finally confirms whether a short circuit has occurred, and estimates the short circuit location after confirmation.
[0030] Communication module 5: This is a 2.4GHz wireless communication chip that sends alarm information to the gateway after confirming a short circuit.
[0031] The device of this invention integrates a data acquisition module, a cache module, a feature extraction module, a processing and judgment module, and a communication module. Through the coordinated operation of each module, it realizes full automation from current signal acquisition, waveform feature analysis, short circuit judgment to alarm information reporting, providing hardware support for quickly locating and eliminating short circuit faults.
[0032] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.
Claims
1. A short-circuit identification method for end-point power consumption data acquisition, characterized in that, include: The current signal in the power supply circuit is acquired in real time and converted into continuous cycle current waveform data; The current waveform data of a continuously preset number of cycles are cached in parallel to form a multi-level cache queue; Waveform features of a consecutive preset number of cycles are synchronously extracted from the multi-level buffer queue; the waveform features include peak clipping features describing current limiting characteristics and shoulder flattening features describing current steady-state characteristics; The waveform features are compared with a preset short-circuit feature rule base over multiple cycles. Based on the consistency of the multi-cycle joint comparison results, it is determined whether a short circuit has occurred.
2. The short-circuit identification method for end-point power consumption data acquisition according to claim 1, characterized in that, The parallel caching of current waveform data for a predetermined number of consecutive cycles specifically includes: Set a preset number of FIFO buffers to be used cyclically; The latest cycle data acquired in real time is stored in the current buffer, and the data in each buffer of the previous cycle is moved to the next cycle in sequence to form the multi-level cache queue.
3. The short-circuit identification method for end-point power consumption data acquisition according to claim 1, characterized in that, Extracting the waveform features specifically includes: Identify the amplitude-limiting segments in the current waveform that exceed a preset current threshold, count them as the number of peak-clipping waveforms, and count the number of sampling points for the duration of the amplitude-limiting segments to obtain the peak-clipping duration period; Identify the steady-state segments in the current waveform that are within a preset current range, count them as the number of flat-shoulder waves, and count the number of sampling points that the steady-state segments last for a period of time to obtain the flat-shoulder wave period.
4. A short-circuit identification method for end-point power consumption data acquisition according to claim 1, characterized in that, Prior to extracting waveform features, the method further includes: Calculate the similarity between the current cycle current waveform data and the pre-stored typical short-circuit waveform; If the similarity is lower than a preset threshold, the current waveform is determined to be a normal load waveform, and the subsequent process is terminated. Otherwise, the step of extracting waveform features is triggered.
5. A short-circuit identification method for end-point power consumption data acquisition according to claim 1, characterized in that, The process of determining whether a short circuit has occurred specifically includes: If the waveform characteristics of the current cycle match the preset short-circuit feature rule base, a pre-alarm status is generated. The pre-alarm status generated by subsequent cycles is compared with the pre-alarm status of previous cycles for consistency. If the alarm status for a consecutive preset number of cycles indicates a short circuit, then a short circuit is confirmed to have occurred.
6. A short-circuit identification method for end-point power consumption data acquisition according to claim 1, characterized in that, Establishing the preset short-circuit feature rule base specifically includes: Short-circuit experiments were conducted at different short-circuit distances, and current waveform data were collected when the short circuit occurred. Statistical analysis of the current waveform data is used to establish a model of the correspondence between short-circuit distance and waveform characteristic quantities.
7. A short-circuit identification method for end-point power consumption data acquisition according to claim 1, characterized in that, After confirming that a short circuit has occurred, the method further includes: Based on the identified flat shoulder wave, its steady-state current value is obtained, and the steady-state current value is determined as the alarm current value for this short circuit event.
8. A short-circuit identification method for end-point power consumption data acquisition according to claim 1, characterized in that, The method further includes: After each short circuit is determined and confirmed on-site, record the waveform characteristics of this short circuit event and the corresponding actual short circuit distance. Based on the newly recorded data, the distance intervals and feature value ranges in the preset short-circuit feature rule base are dynamically corrected.
9. A short-circuit identification device for implementing the method as described in any one of claims 1-8, characterized in that, include: The data acquisition module is used to acquire the current signal in the power supply circuit in real time and convert it into current waveform data; The feature extraction module is used to extract waveform features from the current waveform data. The waveform features include peak clipping features for describing current limiting characteristics and shoulder flattening features for describing current steady-state characteristics. The processing and judgment module is used to compare the extracted waveform features with a preset short-circuit feature rule library, and determine whether a short circuit has occurred based on the comparison result.
10. The short-circuit identification device according to claim 9, characterized in that, The device further includes: The caching module uses multiple buffers to cache the current waveform data, forming a multi-level cache queue; The communication module is used to send the generated alarm information to the gateway via wireless communication after the processing and judgment module determines that a short circuit has occurred. The alarm information includes the alarm current value.