A smart distribution room operation management system and method

Through the intelligent distribution room operation management system, filtering processing and deep learning are used to realize automatic environmental regulation and predictive maintenance of distribution room, which solves the problem of time-consuming, labor-intensive and predictive maintenance of manual inspection in distribution room management, and reduces the problem of insufficient manual inspection and insufficient predictive maintenance in distribution room management. Operation and maintenance and equipment maintenance costs.

CN118336910BActive Publication Date: 2024-11-26上海柘宇电气有限公司
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410435294.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-04-11
Publication Date
2024-11-26
Estimated Expiration
2044-04-11

AI Technical Summary

Technical Problem

The lack of intelligence in the management of distribution rooms has led to time-consuming and labor-intensive manual inspections, and real-time monitoring cannot be achieved, and power outages are prone to occur. There are fewer applications for predictive and preventive maintenance, which increases the cost of operation and maintenance labor and equipment maintenance.

Method used

Adopt the intelligent distribution room operation management system, and generate denoising data through filtering and processing environmental monitoring data. The slide rail patrol robot collects equipment module operation information, and combines deep learning to generate operation abnormal warning models to realize automated environmental regulation and predictive maintenance. , reduce equipment maintenance costs.

Benefits of technology

It realizes automatic environmental regulation and predictive maintenance of the distribution room, improves data accuracy and maintenance efficiency, and reduces operation and maintenance labor and equipment maintenance costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118336910B_ABST
    Figure CN118336910B_ABST
Patent Text Reader

Abstract

The present invention relates to an intelligent distribution room operation management system and method, and belongs to the field of electrical monitoring technology. The present invention includes a data acquisition module, an environmental monitoring module, an equipment monitoring module, a fault prediction module and an early warning control module. De-noised environmental data is obtained through filtering processing according to environmental monitoring data, data deviation is obtained according to the de-noised environmental data and environmental parameter thresholds, and environmental alarm information is generated. The equipment module operation information is collected through a sensor component, and the instrument category and instrument data are obtained through image processing according to the instrument image. The fault level is obtained through operation abnormality early warning model processing according to the equipment module operation information and electrical operation alarm information is generated, and the equipment startup is controlled according to the environmental alarm information, and an operation and maintenance work order is generated according to the electrical operation alarm information and transmitted to a handheld terminal, realizing automatic environmental adjustment, realizing predictive maintenance of the distribution room through intelligent inspection, and reducing equipment maintenance costs.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of electrical monitoring, and in particular relates to an intelligent distribution room operation management system and method. Background Art

[0002] Distribution rooms are characterized by large number, wide distribution, and harsh environment. The monitoring and management of the operating environment of distribution rooms is relatively weak in the operation and maintenance of distribution networks. At present, the management methods of distribution rooms are mostly traditional methods, with less application of artificial intelligence technology and low intelligence of distribution rooms. Traditional management methods are generally unmanned, through manual regular inspections, on-site recording and collection of parameters. Due to the large number and scattered distribution of distribution rooms, there are places that are not convenient for monitoring, which makes manual inspections time-consuming and laborious, and manual inspections cannot achieve real-time monitoring, which is prone to power outages. In addition, the current operation and maintenance of distribution rooms mostly focuses on post-fault processing, while the application of predictive maintenance and preventive maintenance is relatively small, which increases the operation and maintenance manpower costs and equipment maintenance costs. Summary of the invention

[0003] In order to solve the above-mentioned problems existing in the prior art, the present invention provides an intelligent distribution room operation management system and method, which obtains denoised environmental data through filtering according to environmental monitoring data, obtains data deviation according to the denoised environmental data and environmental parameter thresholds, and generates environmental alarm information according to the data deviation. The slide rail inspection robot collects equipment module operation information through sensor components according to preset tracks, obtains instrument categories and instrument data according to instrument images through image processing, obtains operation abnormality warning models through deep learning based on historical equipment module operation information and historical fault information as training sets, obtains fault levels through operation abnormality warning models based on equipment module operation information, generates electrical operation alarm information, controls equipment startup according to environmental alarm information, generates operation and maintenance work orders according to electrical operation alarm information and transmits them to handheld terminals, thereby realizing automatic environmental adjustment, realizing predictive maintenance of distribution rooms through intelligent inspections, and reducing equipment maintenance costs.

[0004] The purpose of the present invention can be achieved through the following technical solutions:

[0005] A method for intelligent power distribution room operation management includes the following steps:

[0006] S1: Collecting environmental monitoring data, wherein the environmental monitoring data includes indoor temperature data and indoor humidity data;

[0007] S2: Obtaining an environmental parameter threshold, obtaining denoised environmental data through filtering according to the environmental monitoring data, obtaining a data deviation according to the denoised environmental data and the environmental parameter threshold, and generating environmental alarm information according to the data deviation, wherein the denoised environmental data includes denoised temperature data and denoised humidity data, and the environmental alarm information includes a temperature abnormality warning and a humidity abnormality warning;

[0008] S3: collecting the equipment module operation information through a preset rail inspection robot, the rail inspection robot carries a sensor component, and the rail inspection robot collects the equipment module operation information through the sensor component according to the preset track, the equipment module operation information includes instrument image information, power distribution cabinet ID, power distribution cabinet temperature information and carbon monoxide concentration information, and the instrument image is processed by image processing to obtain instrument category and instrument data, the instrument category includes voltmeter and ammeter, and the instrument data includes voltage data and current data;

[0009] S4: Obtain historical equipment module operation information and historical fault information, obtain an operation abnormality warning model through deep learning based on the historical equipment module operation information and the historical fault information as a training set, obtain a fault level through the operation abnormality warning model based on the equipment module operation information, and generate electrical operation alarm information based on the fault level in association with the distribution cabinet ID, wherein the fault level includes level one fault, level two fault, and level three fault;

[0010] S5: Control the start-up of the air conditioning equipment according to the abnormal temperature warning, control the start-up of the exhaust device according to the abnormal humidity warning, generate an operation and maintenance work order according to the electrical operation alarm information, set the priority of the operation and maintenance work order according to the fault level, and transmit the operation and maintenance work order to the handheld terminal.

[0011] Preferably, the step S2 specifically includes the following steps:

[0012] S201: Based on the environmental monitoring data, the formula The data anomaly is calculated, where P i Indicates the data abnormality, M i+1 represents the i+1th environmental monitoring data, M i represents the i-th environmental monitoring data, α and β represent weight coefficients;

[0013] S202: Obtain an abnormality threshold value. When the data abnormality is greater than or equal to the data abnormality, the environmental monitoring data corresponding to the data abnormality is recorded as invalid data and deleted. When the data abnormality is less than the abnormality threshold value, the environmental monitoring data corresponding to the data abnormality is recorded as normal data, and the denoised environmental data is obtained according to the normal data.

[0014] S203: The data deviation is calculated according to the denoised environmental data and the environmental parameter threshold value by using the formula AB=AB, wherein AB represents the data deviation, A represents the denoised environmental data, and B represents the environmental parameter threshold value. When AB>0, the environmental alarm information is generated.

[0015] Preferably, it is characterized in that steps S201-S203 are repeated, by adjusting the parameter M i+1 represents the i+1th indoor temperature data, M i Represent the i-th indoor temperature data, obtain the temperature data abnormality, obtain the denoised temperature data according to the temperature data abnormality through the judgment rule of step S202, and obtain the temperature data deviation through step S203 based on the denoised temperature data and the environmental parameter threshold. When the temperature deviation is greater than 0, generate the temperature abnormality warning.

[0016] Preferably, it is characterized in that steps S201-S203 are repeated, by adjusting the parameter M i+1 represents the i+1th indoor humidity data, M i Represent the i-th indoor humidity data, obtain the humidity data abnormality, obtain the denoised humidity data according to the humidity data abnormality through the judgment rule of step S202, calculate the humidity data deviation according to the denoised humidity data through step S203 and the environmental parameter threshold, and generate the humidity abnormality warning when the temperature deviation is greater than 0.

[0017] Preferably, the step S3 specifically further includes the following steps:

[0018] S301: the slide robot identifies a target power distribution cabinet by radio frequency identification technology and obtains the power distribution cabinet ID of the target power distribution cabinet, and the slide robot collects the instrument image of the target power distribution cabinet by self-adjusting its position;

[0019] S302: Acquire standard features of digital display instruments, obtain instrument features through feature extraction based on the instrument image, and obtain feature similarity through cosine similarity calculation based on the standard features of digital display instruments and the instrument features. The calculation formula is: Wherein, A represents the standard feature of the digital instrument, B represents the instrument feature, Sim(A, B) represents the feature similarity, and the instrument category is obtained according to the feature similarity. When the feature similarity is greater than or equal to a preset threshold, the instrument image is a digital instrument image, and when the feature similarity is less than the preset threshold, the instrument image is a pointer instrument image. The instrument corresponding to the digital instrument image is the ammeter, and the instrument corresponding to the pointer instrument image is the voltmeter.

[0020] S303: Using a corresponding recognition algorithm on the instrument image according to the instrument type to obtain the voltage data and the current data.

[0021] Preferably, the step S303 specifically includes the following steps:

[0022] Obtain a target image through image correction according to the digital display instrument image, perform digital display instrument target detection according to the target image to obtain a target digital display area, perform image segmentation on the target area according to the target digital display area to obtain a digital display image, obtain a digital display digital template, perform digital library training through the Tesseract-OCR tool according to the digital display digital template to obtain a digital display character library, and perform template matching on the digital display character library according to the digital display image to obtain the current data;

[0023] The image enhancement data is obtained by processing the pointer instrument image through the image enhancement algorithm, and the formula is: Wherein, K(x, y) represents the image enhancement data, a represents the local linear coefficient, R(x, y) represents the global adaptive output, H(x, y) represents the input image brightness value, L(x, y) represents the Gaussian kernel, (x, y) represents the pixel coordinates, and the instrument enhancement image is obtained according to the image enhancement data. The pointer instrument feature is obtained by feature extraction according to the instrument enhancement image. The instrument pointer and scale line information are obtained by semantic segmentation network processing according to the pointer instrument feature. A coordinate system is established according to the instrument pointer and the scale line information, and the instrument pointer angle is calculated. The calculation formula is: θ=δ+|β|-|θ min |, where θ represents the angle of the instrument pointer, δ represents the coordinate system quadrant coefficient, β represents the angle between the instrument pointer and the x-axis, and θ min Indicates the minimum scale angle, according to the angle of the instrument pointer through the formula The voltage data is obtained by calculation, wherein V represents the voltage data, Vm represents the maximum scale value of the instrument, and Vn represents the minimum scale value of the instrument.

[0024] Preferably, the step S4 specifically includes the following steps:

[0025] According to the distribution cabinet ID, the historical fault feature sequence and the distribution cabinet label information are obtained. According to the equipment module operation information, the equipment module operation feature sequence is obtained by feature extraction. According to the historical fault feature sequence and the equipment module operation feature sequence, the fault probability is obtained by processing the operation abnormality warning model. The calculation formula is: Wherein, pro represents the failure probability, α and β represent weight parameters, [Ltemp, Lco, Lv, La] represents the historical failure feature sequence, temp, co, v and a represent the historical temperature data, historical carbon monoxide concentration, historical voltage data and historical current data of the distribution cabinet failure respectively, [temp, co, v, a] represents the equipment module operation feature sequence, temp represents the temperature information of the distribution cabinet, co represents the carbon monoxide concentration information, v represents the voltage data, a represents the current data, and the distribution cabinet label information is associated with the failure probability through the formula The fault level is calculated, and the distribution cabinet label information includes the distribution cabinet fault frequency and the distribution cabinet level, wherein DJ represents the fault level, T represents the distribution cabinet level, pro represents the fault probability, and Ki represents the equipment failure frequency;

[0026] When 0≤pro<4, the fault level is the primary fault, when 4≤pro<8, the fault level is the secondary fault, and when pro≥8, the fault level is the tertiary fault.

[0027] Preferably, the sensor assembly includes a temperature sensor, a carbon monoxide sensor and a visual sensor:

[0028] An intelligent distribution room operation management system, comprising:

[0029] A data acquisition module is used to collect environmental monitoring data, wherein the environmental monitoring data includes indoor temperature data and indoor humidity data;

[0030] An environmental monitoring module, used to obtain an environmental parameter threshold, obtain denoised environmental data through filtering according to the environmental monitoring data, obtain a data deviation according to the denoised environmental data and the environmental parameter threshold, and generate environmental alarm information according to the data deviation, wherein the denoised environmental data includes denoised temperature data and denoised humidity data, and the environmental alarm information includes a temperature abnormality warning and a humidity abnormality warning;

[0031] An equipment monitoring module is used to collect equipment module operation information through a preset rail inspection robot, the rail inspection robot carries a sensor component, and the rail inspection robot collects the equipment module operation information through the sensor component according to the preset track, the equipment module operation information includes instrument image information, power distribution cabinet ID, power distribution cabinet temperature information and carbon monoxide concentration information, and the instrument image is processed by image processing to obtain instrument category and instrument data, the instrument category includes voltmeter and ammeter, and the instrument data includes voltage data and current data;

[0032] A fault prediction module is used to obtain historical equipment module operation information and historical fault information, obtain an abnormal operation warning model through deep learning based on the historical equipment module operation information and the historical fault information as a training set, obtain a fault level based on the equipment module operation information through the abnormal operation warning model, and generate electrical operation alarm information based on the fault level, wherein the fault level includes primary fault, secondary fault, and tertiary fault;

[0033] The early warning control module is used to control the start-up of the air-conditioning equipment according to the abnormal temperature warning, control the start-up of the exhaust device according to the abnormal humidity warning, generate an operation and maintenance work order according to the electrical operation alarm information, set the priority of the operation and maintenance work order according to the fault level, and transmit the operation and maintenance work order to the handheld terminal.

[0034] The beneficial effects of the present invention are:

[0035] 1. Obtain data anomaly amount according to the environmental monitoring data; when the data anomaly amount is greater than or equal to the data anomaly amount, record the environmental monitoring data corresponding to the data anomaly amount as invalid data and delete it; when the data anomaly amount is less than the anomaly amount threshold, record the environmental monitoring data corresponding to the data anomaly amount as normal data; obtain the denoised environmental data according to the normal data; obtain the data deviation according to the denoised environmental data and the environmental parameter threshold; generate the environmental alarm information according to the data deviation, thereby realizing filtering processing of the environmental monitoring data to obtain the denoised environmental data, improving the accuracy of the data; generating corresponding control instructions according to the environmental alarm information to start the corresponding environmental adjustment device, thereby realizing automatic environmental adjustment of the power distribution room;

[0036] 2. Obtain standard features of digital display instruments, obtain instrument features through feature extraction based on the instrument image, obtain feature similarity through cosine similarity calculation based on the standard features of the digital display instrument and the instrument features, and obtain the instrument category based on the feature similarity. When the feature similarity is greater than or equal to a preset threshold, the instrument image is a digital display instrument image, and when the feature similarity is less than the preset threshold, the instrument image is a pointer-type instrument image. The instrument corresponding to the digital display instrument image is the ammeter, and the instrument corresponding to the pointer-type instrument image is the voltmeter. By distinguishing the voltmeter and the ammeter using the pointer type and the digital display type, the purpose is to recognize the instrument image through the image during the inspection and recognition process. According to the difference in instrument categories, voltage data and current data are obtained by setting different recognition algorithms, thereby improving image processing speed and data recognition accuracy, and avoiding data errors caused by instrument confusion.

[0037] 3. By acquiring historical equipment module operation information and historical fault information, an operation abnormality warning model is obtained through deep learning based on the historical equipment module operation information and the historical fault information as a training set, and an equipment module operation feature sequence is obtained through feature extraction based on the equipment module operation information. The fault probability is obtained through abnormal warning model processing based on the historical fault feature sequence and the equipment module operation feature sequence, thereby realizing the prediction of distribution room operation abnormalities, correlating the distribution cabinet label information according to the fault probability to obtain the fault level, setting the priority of the operation and maintenance work order according to the fault level, and transmitting the operation and maintenance work order to the handheld terminal, thereby realizing predictive maintenance of the distribution room, and arranging manual troubleshooting and maintenance by predicting the probability of fault occurrence, thereby reducing equipment maintenance costs and operation and maintenance manpower costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0039] Figure 1 It is a flow chart of the intelligent power distribution room operation management method of the present invention. DETAILED DESCRIPTION

[0040] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0041] See also Figure 1 , a smart distribution room operation management method, specifically comprising the following steps:

[0042] S1: Collecting environmental monitoring data, wherein the environmental monitoring data includes indoor temperature data and indoor humidity data;

[0043] S2: Obtaining an environmental parameter threshold, obtaining denoised environmental data through filtering according to the environmental monitoring data, obtaining a data deviation according to the denoised environmental data and the environmental parameter threshold, and generating environmental alarm information according to the data deviation, wherein the denoised environmental data includes denoised temperature data and denoised humidity data, and the environmental alarm information includes a temperature abnormality warning and a humidity abnormality warning;

[0044] S3: collecting the equipment module operation information through a preset rail inspection robot, the rail inspection robot carries a sensor component, and the rail inspection robot collects the equipment module operation information through the sensor component according to the preset track, the equipment module operation information includes instrument image information, power distribution cabinet ID, power distribution cabinet temperature information and carbon monoxide concentration information, and the instrument image is processed by image processing to obtain instrument category and instrument data, the instrument category includes voltmeter and ammeter, and the instrument data includes voltage data and current data;

[0045] S4: Obtain historical equipment module operation information and historical fault information, obtain an operation abnormality warning model through deep learning based on the historical equipment module operation information and the historical fault information as a training set, obtain a fault level through the operation abnormality warning model based on the equipment module operation information, and generate electrical operation alarm information based on the fault level in association with the distribution cabinet ID, wherein the fault level includes level one fault, level two fault, and level three fault;

[0046] S5: Control the start-up of the air conditioning equipment according to the abnormal temperature warning, control the start-up of the exhaust device according to the abnormal humidity warning, generate an operation and maintenance work order according to the electrical operation alarm information, set the priority of the operation and maintenance work order according to the fault level, and transmit the operation and maintenance work order to the handheld terminal.

[0047] Specifically, the step S2 comprises the following steps:

[0048] S201: Based on the environmental monitoring data, the formula The data anomaly is calculated, where P i Indicates the data abnormality, M i+1 represents the i+1th environmental monitoring data, M i represents the i-th environmental monitoring data, α and β represent weight coefficients;

[0049] S202: Obtain an abnormality threshold value. When the data abnormality is greater than or equal to the data abnormality, the environmental monitoring data corresponding to the data abnormality is recorded as invalid data and deleted. When the data abnormality is less than the abnormality threshold value, the environmental monitoring data corresponding to the data abnormality is recorded as normal data, and the denoised environmental data is obtained according to the normal data.

[0050] S203: The data deviation is calculated according to the denoised environmental data and the environmental parameter threshold value by using the formula AB=AB, wherein AB represents the data deviation, A represents the denoised environmental data, and B represents the environmental parameter threshold value. When AB>0, the environmental alarm information is generated.

[0051] Specifically, it is characterized in that steps S201-S203 are repeated, by adjusting the parameter M i+1 represents the i+1th indoor temperature data, M i Represent the i-th indoor temperature data, obtain the temperature data abnormality, obtain the denoised temperature data according to the temperature data abnormality through the judgment rule of step S202, and obtain the temperature data deviation through step S203 based on the denoised temperature data and the environmental parameter threshold. When the temperature deviation is greater than 0, generate the temperature abnormality warning.

[0052] Specifically, it is characterized in that steps S201-S203 are repeated, by adjusting the parameter M i+1 represents the i+1th indoor humidity data, M i Represent the i-th indoor humidity data, obtain the humidity data abnormality, obtain the denoised humidity data according to the humidity data abnormality through the judgment rule of step S202, calculate the humidity data deviation according to the denoised humidity data through step S203 and the environmental parameter threshold, and generate the humidity abnormality warning when the temperature deviation is greater than 0.

[0053] Specifically, the step S3 further includes the following steps:

[0054] S301: the slide robot identifies a target power distribution cabinet by radio frequency identification technology and obtains the power distribution cabinet ID of the target power distribution cabinet, and the slide robot collects the instrument image of the target power distribution cabinet by self-adjusting its position;

[0055] S302: Acquire standard features of digital display instruments, obtain instrument features through feature extraction based on the instrument image, and obtain feature similarity through cosine similarity calculation based on the standard features of digital display instruments and the instrument features. The calculation formula is: Wherein, A represents the standard feature of the digital instrument, B represents the instrument feature, Sim(A, B) represents the feature similarity, and the instrument category is obtained according to the feature similarity. When the feature similarity is greater than or equal to a preset threshold, the instrument image is a digital instrument image, and when the feature similarity is less than the preset threshold, the instrument image is a pointer instrument image. The instrument corresponding to the digital instrument image is the ammeter, and the instrument corresponding to the pointer instrument image is the voltmeter.

[0056] By distinguishing between the pointer type and the digital display type for voltmeters and ammeters, it is possible to recognize instrument images through images during the inspection and identification process. According to the difference in instrument categories, voltage data and current data are obtained by setting different recognition algorithms, thereby improving image processing speed and data recognition accuracy and avoiding data errors caused by instrument confusion.

[0057] S303: Using a corresponding recognition algorithm on the instrument image according to the instrument type to obtain the voltage data and the current data.

[0058] Specifically, step S303 includes the following steps:

[0059] Obtain a target image through image correction according to the digital display instrument image, perform digital display instrument target detection according to the target image to obtain a target digital display area, perform image segmentation on the target area according to the target digital display area to obtain a digital display image, obtain a digital display digital template, perform digital library training through the Tesseract-OCR tool according to the digital display digital template to obtain a digital display character library, and perform template matching on the digital display character library according to the digital display image to obtain the current data;

[0060] The image enhancement data is obtained by processing the pointer instrument image through the image enhancement algorithm, and the formula is: Wherein, K(x, y) represents the image enhancement data, a represents the local linear coefficient, R(x, y) represents the global adaptive output, H(x, y) represents the input image brightness value, L(x, y) represents the Gaussian kernel, (x, y) represents the pixel coordinates, and the instrument enhancement image is obtained according to the image enhancement data. The pointer instrument feature is obtained by feature extraction according to the instrument enhancement image. The instrument pointer and scale line information are obtained by semantic segmentation network processing according to the pointer instrument feature. A coordinate system is established according to the instrument pointer and the scale line information, and the instrument pointer angle is calculated. The calculation formula is: θ=δ+|β|-|θ min |, where θ represents the angle of the instrument pointer, δ represents the coordinate system quadrant coefficient, β represents the angle between the instrument pointer and the x-axis, and θ min Indicates the minimum scale angle, according to the angle of the instrument pointer through the formula The voltage data is obtained by calculation, wherein V represents the voltage data, Vm represents the maximum scale value of the instrument, and Vn represents the minimum scale value of the instrument.

[0061] Specifically, the step S4 comprises the following steps:

[0062] According to the distribution cabinet ID, the historical fault feature sequence and the distribution cabinet label information are obtained. According to the equipment module operation information, the equipment module operation feature sequence is obtained by feature extraction. According to the historical fault feature sequence and the equipment module operation feature sequence, the fault probability is obtained by processing the operation abnormality warning model. The calculation formula is: Wherein, pro represents the failure probability, α and β represent weight parameters, [Ltemp, Lco, Lv, La] represents the historical failure feature sequence, temp, co, v and a represent the historical temperature data, historical carbon monoxide concentration, historical voltage data and historical current data of the distribution cabinet failure respectively, [temp, co, v, a] represents the equipment module operation feature sequence, temp represents the temperature information of the distribution cabinet, co represents the carbon monoxide concentration information, v represents the voltage data, a represents the current data, and the distribution cabinet label information is associated with the failure probability through the formula The fault level is calculated, and the distribution cabinet label information includes the distribution cabinet fault frequency and the distribution cabinet level, wherein DJ represents the fault level, T represents the distribution cabinet level, pro represents the fault probability, and Ki represents the equipment failure frequency;

[0063] When 0≤pro<4, the fault level is the primary fault, when 4≤pro<8, the fault level is the secondary fault, and when pro≥8, the fault level is the tertiary fault.

[0064] By acquiring historical equipment module operation information and historical fault information, an operation abnormality warning model is obtained through deep learning based on the historical equipment module operation information and the historical fault information as a training set, an equipment module operation feature sequence is obtained through feature extraction based on the equipment module operation information, and the fault probability is obtained through abnormal warning model processing based on the historical fault feature sequence and the equipment module operation feature sequence, thereby realizing the prediction of distribution room operation abnormalities, correlating distribution cabinet label information according to the fault probability to obtain the fault level, setting the priority of the operation and maintenance work order according to the fault level, and transmitting the operation and maintenance work order to the handheld terminal, thereby realizing predictive maintenance of the distribution room, and arranging manual troubleshooting and maintenance by predicting the probability of fault occurrence, thereby reducing equipment maintenance costs and operation and maintenance manpower costs.

[0065] Specifically, the sensor assembly includes a temperature sensor, a carbon monoxide sensor and a visual sensor:

[0066] Furthermore, the present invention also provides an intelligent distribution room operation management system, comprising:

[0067] A data acquisition module is used to collect environmental monitoring data, wherein the environmental monitoring data includes indoor temperature data and indoor humidity data;

[0068] An environmental monitoring module, used to obtain an environmental parameter threshold, obtain denoised environmental data through filtering according to the environmental monitoring data, obtain a data deviation according to the denoised environmental data and the environmental parameter threshold, and generate environmental alarm information according to the data deviation, wherein the denoised environmental data includes denoised temperature data and denoised humidity data, and the environmental alarm information includes a temperature abnormality warning and a humidity abnormality warning;

[0069] An equipment monitoring module is used to collect equipment module operation information through a preset rail inspection robot, the rail inspection robot carries a sensor component, and the rail inspection robot collects the equipment module operation information through the sensor component according to the preset track, the equipment module operation information includes instrument image information, power distribution cabinet ID, power distribution cabinet temperature information and carbon monoxide concentration information, and the instrument image is processed by image processing to obtain instrument category and instrument data, the instrument category includes voltmeter and ammeter, and the instrument data includes voltage data and current data;

[0070] A fault prediction module is used to obtain historical equipment module operation information and historical fault information, obtain an abnormal operation warning model through deep learning based on the historical equipment module operation information and the historical fault information as a training set, obtain a fault level based on the equipment module operation information through the abnormal operation warning model, and generate electrical operation alarm information based on the fault level, wherein the fault level includes primary fault, secondary fault, and tertiary fault;

[0071] The early warning control module is used to control the start-up of the air-conditioning equipment according to the abnormal temperature warning, control the start-up of the exhaust device according to the abnormal humidity warning, generate an operation and maintenance work order according to the electrical operation alarm information, set the priority of the operation and maintenance work order according to the fault level, and transmit the operation and maintenance work order to the handheld terminal.

[0072] The working principle and use process of the present invention:

[0073] Environmental monitoring data is collected through the data acquisition module, and denoised environmental data is obtained through filtering based on the environmental monitoring data. Data deviation is obtained based on the denoised environmental data and environmental parameter thresholds, and environmental alarm information is generated based on the data deviation. The equipment module operation information is collected through the preset slide rail inspection robot. The slide rail inspection robot carries a sensor component. The slide rail inspection robot collects the equipment module operation information through the sensor component according to the preset track, and obtains the instrument category and instrument data through image processing of the instrument image. The operation abnormality warning model is obtained through deep learning based on the historical equipment module operation information and historical fault information as a training set. The fault level is obtained by processing the equipment module operation information through the operation abnormality warning model, and the electrical operation alarm information is generated by associating the distribution cabinet ID according to the fault level. The air conditioning equipment is controlled to start according to the temperature abnormality warning, and the exhaust device is controlled to start according to the humidity abnormality warning. An operation and maintenance work order is generated according to the electrical operation alarm information, and the operation and maintenance work order priority is set according to the fault level, and the operation and maintenance work order is transmitted to the handheld terminal.

[0074] The program code included in the method in the embodiment of the present invention can be transmitted with any appropriate medium, including but not limited to wireless, electric wire, optical cable, etc., or any suitable combination mentioned above. The computer program code for performing the operation of the present invention can be written with one or more programming languages ​​or their combination, and the programming language includes object-oriented programming languages--such as Java, Smalltalk, C++, and also includes conventional procedural programming languages--such as "C" language or similar programming languages. The program code can be executed completely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on the remote computer, or executed completely on the remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0075] The above is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still belong to the specification of the technical solution of the present invention.

Claims

1. A smart distribution room operation management method, It is characterized in that The specific steps include: S1: Collecting environmental monitoring data, wherein the environmental monitoring data includes indoor temperature data and indoor humidity data; S2: Obtaining an environmental parameter threshold, obtaining denoised environmental data through filtering according to the environmental monitoring data, obtaining a data deviation according to the denoised environmental data and the environmental parameter threshold, and generating environmental alarm information according to the data deviation, wherein the denoised environmental data includes denoised temperature data and denoised humidity data, and the environmental alarm information includes a temperature abnormality warning and a humidity abnormality warning; S3: collecting the equipment module operation information through a preset rail inspection robot, the rail inspection robot carries a sensor component, and the rail inspection robot collects the equipment module operation information through the sensor component according to the preset track, the equipment module operation information includes instrument image information, power distribution cabinet ID, power distribution cabinet temperature information and carbon monoxide concentration information, and the instrument image is processed by image processing to obtain instrument category and instrument data, the instrument category includes voltmeter and ammeter, and the instrument data includes voltage data and current data; S4: Obtain historical equipment module operation information and historical fault information, obtain an abnormal operation warning model through deep learning based on the historical equipment module operation information and the historical fault information as a training set, obtain a fault level through the abnormal operation warning model based on the equipment module operation information, and generate electrical operation alarm information based on the fault level in association with the distribution cabinet ID. The fault level includes a primary fault, a secondary fault, and a tertiary fault. The fault level obtained through the abnormal operation warning model based on the equipment module operation information is specifically: According to the distribution cabinet ID, the historical fault feature sequence and the distribution cabinet label information are obtained. According to the equipment module operation information, the equipment module operation feature sequence is obtained by feature extraction. According to the historical fault feature sequence and the equipment module operation feature sequence, the fault probability is obtained by processing the operation abnormality warning model. The calculation formula is: Wherein, pro represents the failure probability, α and β represent weight parameters, [Ltemp, Lco, Lv, La] represents the historical failure feature sequence, temp, co, v and a represent the historical temperature data, historical carbon monoxide concentration, historical voltage data and historical current data of the distribution cabinet failure respectively, [temp, co, v, a] represents the equipment module operation feature sequence, temp represents the temperature information of the distribution cabinet, co represents the carbon monoxide concentration information, v represents the voltage data, a represents the current data, and the distribution cabinet label information is associated with the failure probability through the formula The fault level is calculated, and the distribution cabinet label information includes the distribution cabinet fault frequency and the distribution cabinet level, wherein DJ represents the fault level, T represents the distribution cabinet level, pro represents the fault probability, and Ki represents the equipment failure frequency; When 0≤pro<4, the fault level is the first-level fault, when 4≤pro<8, the fault level is the second-level fault, and when pro≥8, the fault level is the third-level fault; S5: Control the start-up of the air conditioning equipment according to the abnormal temperature warning, control the start-up of the exhaust device according to the abnormal humidity warning, generate an operation and maintenance work order according to the electrical operation alarm information, set the priority of the operation and maintenance work order according to the fault level, and transmit the operation and maintenance work order to the handheld terminal.

2. According to the intelligent distribution room operation management method of claim 1, It is characterized in that The step S2 specifically includes the following steps: S201: Based on the environmental monitoring data, the formula The data anomaly is calculated, where P i Indicates the data abnormality, M i+1 represents the i+1th environmental monitoring data, M i represents the i-th environmental monitoring data, α and β represent weight coefficients; S202: Obtain an abnormality threshold value. When the data abnormality is greater than or equal to the data abnormality, the environmental monitoring data corresponding to the data abnormality is recorded as invalid data and deleted. When the data abnormality is less than the abnormality threshold value, the environmental monitoring data corresponding to the data abnormality is recorded as normal data, and the denoised environmental data is obtained according to the normal data. S203: The data deviation is calculated according to the denoised environmental data and the environmental parameter threshold value by using the formula AB=AB, wherein AB represents the data deviation, A represents the denoised environmental data, and B represents the environmental parameter threshold value. When AB>0, the environmental alarm information is generated.

3. According to the intelligent distribution room operation management method of claim 2, It is characterized in that Repeat steps S201-S203, and adjust the parameter M i+1 represents the i+1th indoor temperature data, M i Representing the i-th indoor temperature data, obtaining the temperature data abnormality, obtaining the denoised temperature data according to the temperature data abnormality through the judgment rule of step S202, obtaining the temperature data deviation through step S203 and the environmental parameter threshold according to the denoised temperature data, and generating the temperature abnormality warning when the temperature data deviation is greater than 0.

4. According to claim 2, the intelligent power distribution room operation management method, It is characterized in that Repeat steps S201-S203, and adjust the parameter M i+1 represents the i+1th indoor humidity data, M i Represent the i-th indoor humidity data, obtain the humidity data abnormality, obtain the denoised humidity data according to the humidity data abnormality through the judgment rule of step S202, and obtain the humidity data deviation through step S203 and the environmental parameter threshold according to the denoised humidity data. When the humidity data deviation is greater than 0, generate the humidity abnormality warning.

5. According to the intelligent distribution room operation management method of claim 1, It is characterized in that The step S3 specifically further comprises the following steps: S301: the rail inspection robot identifies a target power distribution cabinet by radio frequency identification technology and obtains the power distribution cabinet ID of the target power distribution cabinet, and the rail inspection robot collects the instrument image of the target power distribution cabinet by self-adjusting its position; S302: Obtain standard features of digital display instruments, obtain instrument features through feature extraction based on the instrument image, and obtain feature similarity through cosine similarity calculation based on the standard features of the digital display instrument and the instrument features. The calculation formula is: Wherein, A represents the standard feature of the digital instrument, B represents the instrument feature, Sim(A,B) represents the feature similarity, and the instrument category is obtained according to the feature similarity. When the feature similarity is greater than or equal to a preset threshold, the instrument image is a digital instrument image, and when the feature similarity is less than the preset threshold, the instrument image is a pointer instrument image. The instrument corresponding to the digital instrument image is the ammeter, and the instrument corresponding to the pointer instrument image is the voltmeter. S303: Using a corresponding recognition algorithm on the instrument image according to the instrument type to obtain the voltage data and the current data.

6. According to the intelligent distribution room operation management method of claim 5, It is characterized in that The step S303 specifically includes the following steps: Obtain a target image through image correction according to the digital display instrument image, perform digital display instrument target detection according to the target image to obtain a target digital display area, perform image segmentation on the target area according to the target digital display area to obtain a digital display image, obtain a digital display digital template, perform digital library training through the Tesseract-OCR tool according to the digital display digital template to obtain a digital display character library, and perform template matching on the digital display character library according to the digital display image to obtain the current data; The image enhancement data is obtained by processing the pointer instrument image through the image enhancement algorithm, and the formula is: Wherein, K(x, y) represents the image enhancement data, a represents the local linear coefficient, R(x, y) represents the global adaptive output, H(x, y) represents the input image brightness value, L(x, y) represents the Gaussian kernel, (x, y) represents the pixel coordinates, and the instrument enhancement image is obtained according to the image enhancement data. The pointer instrument feature is obtained by feature extraction according to the instrument enhancement image. The instrument pointer and scale line information are obtained by semantic segmentation network processing according to the pointer instrument feature. A coordinate system is established according to the instrument pointer and the scale line information, and the instrument pointer angle is calculated. The calculation formula is: θ=δ+|β|-|θ min |, where θ represents the angle of the instrument pointer, δ represents the coordinate system quadrant coefficient, β represents the angle between the instrument pointer and the x-axis, and θ min Indicates the minimum scale angle, according to the angle of the instrument pointer through the formula The voltage data is obtained by calculation, wherein V represents the voltage data, Vm represents the maximum scale value of the instrument, and Vn represents the minimum scale value of the instrument.

7. According to claim 1, the intelligent power distribution room operation management method, It is characterized in that The sensor assembly includes a temperature sensor, a carbon monoxide sensor, and a visual sensor.

8. An intelligent power distribution room operation management system, which adopts the intelligent power distribution room operation management method as described in any one of claims 1 to 7, It is characterized in that include: A data acquisition module is used to collect environmental monitoring data, wherein the environmental monitoring data includes indoor temperature data and indoor humidity data; An environmental monitoring module, used to obtain an environmental parameter threshold, obtain denoised environmental data through filtering according to the environmental monitoring data, obtain a data deviation according to the denoised environmental data and the environmental parameter threshold, and generate environmental alarm information according to the data deviation, wherein the denoised environmental data includes denoised temperature data and denoised humidity data, and the environmental alarm information includes a temperature abnormality warning and a humidity abnormality warning; An equipment monitoring module is used to collect equipment module operation information through a preset rail inspection robot, the rail inspection robot carries a sensor component, and the rail inspection robot collects the equipment module operation information through the sensor component according to the preset track, the equipment module operation information includes instrument image information, power distribution cabinet ID, power distribution cabinet temperature information and carbon monoxide concentration information, and the instrument image is processed by image processing to obtain instrument category and instrument data, the instrument category includes voltmeter and ammeter, and the instrument data includes voltage data and current data; A fault prediction module is used to obtain historical equipment module operation information and historical fault information, obtain an abnormal operation warning model through deep learning based on the historical equipment module operation information and the historical fault information as a training set, obtain a fault level based on the equipment module operation information through the abnormal operation warning model, and generate electrical operation alarm information based on the fault level, wherein the fault level includes primary fault, secondary fault, and tertiary fault; The early warning control module is used to control the start-up of the air-conditioning equipment according to the abnormal temperature warning, control the start-up of the exhaust device according to the abnormal humidity warning, generate an operation and maintenance work order according to the electrical operation alarm information, set the priority of the operation and maintenance work order according to the fault level, and transmit the operation and maintenance work order to the handheld terminal.