Intelligent park management method and system based on artificial intelligence large model
By comparing and analyzing the energy consumption data of smart park equipment using a large artificial intelligence model, the problem of delayed judgment of abnormal power consumption was solved, enabling timely maintenance and energy-saving management of equipment.
Patent Information
- Application Number
- CN202511201010.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-26
- Publication Date
- 2025-10-28
AI Technical Summary
In existing smart campus equipment management methods, there is a lag in judging abnormal power consumption, resulting in energy waste and equipment damage or safety risks.
A smart park management method based on a large artificial intelligence model is adopted. By obtaining the reference energy consumption data of the equipment and comparing it with the real-time monitoring data, the comparison model and energy consumption dispersion model are used to judge the abnormal power consumption of the equipment, obtain the starting time point of the energy consumption anomaly, and dynamically adjust the inspection strategy.
It improves the accuracy and timeliness of abnormal power consumption detection, ensures the stability and energy efficiency of equipment operation, and promptly identifies and addresses equipment risks.
Smart Images
Figure CN120850171A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of smart park management technology, specifically to a smart park management method and system based on a large-scale artificial intelligence model. Background Art
[0002] With the development and implementation of IoT and AI technologies, they have provided significant assistance in park management. By setting up sensing facilities at the edge and relying on the park's smart platform, the operational status within the park can be monitored in real time, facilitating effective park management.
[0003] In the process of managing park equipment, existing technologies can acquire real-time energy consumption data by monitoring equipment operation data. By analyzing the energy consumption data, it is possible to promptly identify abnormal energy consumption, thus avoiding energy waste and achieving energy-saving effects. On the other hand, timely identification and handling of power-consuming equipment can prevent the exaggeration of equipment problems, explain the impact of equipment failures, and prevent safety issues from occurring.
[0004] While existing smart park equipment management methods can identify equipment problems, there is a certain lag in the process of identifying abnormal power consumption. That is, the system can only identify abnormal power consumption when the power consumption of equipment significantly exceeds its average range. On the one hand, the lag in identifying abnormal power consumption leads to some energy waste. More importantly, the inability to identify abnormal power consumption in a timely manner can lead to equipment damage or safety risks. Therefore, how to more timely and accurately identify abnormal power consumption of park equipment is the fundamental problem that this invention aims to solve. Summary of the Invention
[0005] The purpose of this invention is to provide a smart park management method and system based on a large-scale artificial intelligence model, and to solve the following technical problems: How to more promptly and accurately identify abnormal power consumption of equipment in the park.
[0006] The purpose of the present invention can be achieved through the following technical solutions: A smart park management method based on an artificial intelligence big data model, the method comprising: The environmental parameters and equipment operating parameters within the park are acquired and input into the large-scale artificial intelligence model to obtain reference energy consumption data for each device; The reference energy consumption data of each device is compared with the real-time monitored energy consumption data to obtain the first comparison result. The first comparison results of the same type of devices are then compared to obtain the second comparison result. Based on the first comparison result and the second comparison result, it is determined whether there is any abnormality in the energy consumption status of the device. When an abnormal energy consumption status of a device is detected, the starting time of the abnormal energy consumption is obtained, and the risk of the device is assessed based on the starting time of the abnormal energy consumption and the device's monitoring data.
[0007] The above technical solutions enable more accurate and timely judgment of abnormal power consumption in equipment, thereby improving the accuracy and timeliness of the judgment. This facilitates timely maintenance of equipment by park management personnel based on the judgment results, ensuring the stability and energy efficiency of equipment operation.
[0008] Furthermore, the process of acquiring reference energy consumption data for each device includes: The environmental parameters and operating parameters of each device within an analysis period are obtained and input into the large artificial intelligence model to obtain the reference power consumption of the device within the analysis period. The reference power consumption of the continuous analysis cycle is placed in a plane coordinate system, and the points in the coordinate system are fitted to obtain the reference power consumption curve of each device. The reference power consumption curve is used as the reference energy consumption data.
[0009] The above technical solution can determine the changes in data at different time points, thus improving the accuracy of the first comparison result during the comparison process.
[0010] Furthermore, the first comparison process includes: The real-time monitored energy consumption data is fitted to a real-time power consumption curve E(t). The real-time power consumption curve E(t) and the reference power consumption curve Er(t) are placed in the same coordinate system, and a comparison model is constructed to obtain the deviation dev. The comparison model is expressed as follows:
[0011] Where tx is the start time point and ty is the end time point; Compare the deviation amount dev with the preset error tolerance: When the deviation dev exceeds the preset error allowable amount, it is determined that there is an abnormality in the energy consumption status; A second comparison is performed when the deviation dev is less than or equal to the preset error allowable amount.
[0012] The above technical solution enables more accurate judgments based on data deviations at different time periods, thereby improving the accuracy of the judgments.
[0013] Furthermore, the second comparison process includes: An energy consumption dispersion model is constructed for each device, and the deviation consistency coefficient s of each device is obtained. The energy consumption dispersion model is represented as follows:
[0014] Where n is the number of devices in this category, and i is a positive integer and i∈[1,n]. Let be the deviation of the i-th device. Let i be the energy consumption factor of the i-th device. For all devices in this category The mean; Compare the deviation consistency coefficient s with the preset threshold st: If s≤st, the reference power consumption of all devices is summed and compared with the total power consumption. If the comparison error exceeds the preset error threshold, it is determined that there is a risk of leakage; otherwise, it is determined that the device is operating normally. If s > st, according to Sort the devices from largest to smallest, exclude the data corresponding to the first device in the sort, recalculate the deviation consistency coefficient s, and repeat this process until s≤st. Then, identify the devices whose data has been excluded as risk devices.
[0015] The above technical solution can eliminate the influence of differences in energy consumption ratios of different devices on the comparison of device deviations, thereby improving the accuracy of the judgment results and realizing the judgment process for devices with abnormal power consumption.
[0016] Furthermore, the process of obtaining the starting time point of energy consumption anomalies includes: A model for determining the starting time point of energy consumption anomalies is constructed to obtain the starting time point tp of energy consumption anomalies. The model for determining the starting time point of energy consumption anomalies is represented as follows:
[0017] in, express The maximum value corresponds to the time point. Preset fixed time difference.
[0018] The above technical solutions can assist in a more accurate analysis and judgment of the equipment status.
[0019] Furthermore, the process of assessing equipment risk includes: Determine the equipment's operating status after the initial time point of abnormal energy consumption based on the equipment's operating parameters: If the device is not in operation, it is determined that the device has been abnormally activated; If the equipment is in operation, acquire its temperature data, construct a temperature analysis model based on different equipment types, and identify temperature anomaly factors. The temperature analysis model is expressed as follows:
[0020] Where T(t) is the equipment temperature change curve, and Te(t) is the ambient temperature change curve. For preset time periods; When temperature anomaly factor If the value exceeds the preset value, the device is considered to have a hardware malfunction; otherwise, the device is considered to have an abnormal configuration.
[0021] The above technical solutions can help determine the specific reasons for abnormal power consumption of equipment, thereby facilitating timely adaptive maintenance of the equipment by park management personnel.
[0022] Furthermore, the method also includes: The patrol vehicle patrol strategy is dynamically adjusted based on the first and second comparison results, including: Adjust the value of the preset threshold st to sr, and satisfy sr < st. Select the devices that exclude data under the condition s ≤ sr as the predicted risk devices, obtain the location points of the predicted risk devices, and increase the frequency of patrol vehicles patrolling the location points of the predicted risk devices.
[0023] The above technical solutions can increase the frequency of patrols by patrol vehicles at the locations of the predicted risk equipment, ensuring that safety risks to the equipment can be dealt with in a timely manner.
[0024] A smart park management system based on an artificial intelligence big data model, the system being used to execute a smart park management method based on an artificial intelligence big data model as described above, including an energy consumption data acquisition terminal, an equipment monitoring terminal, and a processor; The energy consumption data acquisition terminal is used to acquire environmental parameters and equipment operating parameters within the park and input them into the artificial intelligence big model to obtain reference energy consumption data for each device; The device monitoring terminal is used to acquire monitoring data from the device. The processor is used to perform a first comparison between the reference energy consumption data of each device and the real-time monitored energy consumption data to obtain a first comparison result, and to perform a second comparison between the first comparison results of devices of the same category to obtain a second comparison result. Based on the first comparison result and the second comparison result, it is determined whether there is an abnormality in the energy consumption status of the device. When it is determined that the energy consumption status of the device is abnormal, the processor obtains the start time point of the energy consumption abnormality, and judges the risk of the device based on the start time point of the energy consumption abnormality and the monitoring data of the device.
[0025] The beneficial effects of this invention are: (1) By comparing the reference energy consumption data of each device with the real-time monitored energy consumption data, the present invention can more accurately and timely judge the abnormal power consumption of the device; by comparing the first comparison results of the same type of device with the second comparison, the abnormal power consumption of the device can be further judged, thereby improving the accuracy and timeliness of the judgment of abnormal power consumption of the device; by obtaining the starting time point of the abnormal power consumption, the device risk can be judged according to the starting time point of the abnormal power consumption and the monitoring data of the device, thereby enabling timely auxiliary judgment of the cause of abnormal power consumption, which facilitates the park management personnel to maintain the device in a timely manner according to the judgment results, and ensures the stability and energy saving of the device operation. Attached Figure Description
[0026] The invention will now be further described with reference to the accompanying drawings.
[0027] Figure 1 This is a flowchart illustrating the steps of a smart park management method based on a large artificial intelligence model according to the present invention. Detailed Implementation
[0028] 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 embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0029] In one embodiment, a smart park management method based on an artificial intelligence big data model is provided. Please refer to [link / reference]. Figure 1As shown, the method includes: acquiring environmental parameters and equipment operating parameters within the park and inputting them into an artificial intelligence (AI) big data model to obtain reference energy consumption data for each device. The AI big data model is trained using historical operating parameter data and power consumption data for each device as sample data. The training process includes data preprocessing, feature selection, training based on a deep learning model (LSTM), model evaluation, and model optimization, which will not be detailed here. Through the AI big data model, accurate reference energy consumption data can be obtained as a comparison reference. Then, the reference energy consumption data for each device is compared with the real-time monitored energy consumption data to obtain the first comparison result. Compared to the existing method of comparing energy consumption data with the device's power consumption warning value, the obtained reference energy consumption data better reflects the true power consumption status. Therefore, by comparing the reference energy consumption data for each device with the real-time monitored energy consumption data... This process allows for more accurate and timely judgment of abnormal equipment power consumption. Then, a second comparison is made with the first comparison results of similar equipment to obtain the second comparison results. Based on the first and second comparison results, it is determined whether there are any abnormalities in the equipment's energy consumption status. Building upon the first comparison, since the equipment's power consumption status is affected by various factors, a second comparison is made with the first comparison results of similar equipment to further judge abnormal power consumption, thereby improving the accuracy and timeliness of the judgment. Furthermore, when judging abnormal equipment energy consumption, the starting time point of the energy consumption anomaly is obtained. Based on the starting time point and the equipment's monitoring data, the risk of the equipment is judged, which allows for timely auxiliary judgment of the cause of the abnormal power consumption. This facilitates timely maintenance of the equipment by park management personnel based on the judgment results, ensuring the stability and energy efficiency of equipment operation.
[0030] In one embodiment, the process of acquiring reference energy consumption data for each device includes: acquiring environmental parameters and operating parameters of each device within an analysis period and inputting them into an artificial intelligence model to obtain the reference power consumption of the device within the analysis period; placing the reference power consumption of continuous analysis periods in a planar coordinate system, fitting the points in the coordinate system to obtain the reference power consumption curve of each device, and using the reference power consumption curve as reference energy consumption data. Since the result predicted by the artificial intelligence model is energy consumption data over a period of time, although the obtained result has high accuracy over an overall period, it cannot reflect the changing state of the data. However, this embodiment, by dividing the analysis period, acquires environmental parameters and operating parameters of each device within an analysis period and inputs them into an artificial intelligence model to obtain the reference power consumption of the device within the analysis period, and then obtains the reference power consumption curve of each device through fitting. This method can determine the changing state of the data at different time points, thus improving the accuracy of the first comparison result during the comparison process.
[0031] In one embodiment, a first comparison process is provided, comprising: firstly fitting the real-time monitored energy consumption data into a real-time power consumption curve E(t); then placing the real-time power consumption curve E(t) and the reference power consumption curve Er(t) in the same coordinate system to construct a comparison model to obtain the deviation dev, wherein the comparison model is represented as follows:
[0032] Where tx is the start time and ty is the end time; the time period from the start time to the end time is set by the user, for example, using one full hour as the start time and the next full hour as the end time. The deviation within one hour is compared to determine the abnormal power consumption state. The deviation dev is compared with the preset error allowable value, which is set based on experience data of different types of equipment. Therefore, when the deviation dev is greater than the preset error allowable value, it indicates that the deviation is significantly too large, and thus the power consumption state is judged to be abnormal. When the deviation dev is less than or equal to the preset error allowable value, a second comparison is performed for further judgment. Through the above comparison model, the judgment can be made more accurately based on the data deviation in different time periods, thus improving the accuracy of the judgment.
[0033] In one embodiment, a second comparison process is provided, including: constructing an energy consumption dispersion model for each device, obtaining the deviation consistency coefficient s for each device, wherein the energy consumption dispersion model is represented as follows:
[0034] Where n is the number of devices in this category, and i is a positive integer and i∈[1,n]. Let be the deviation of the i-th device. For all devices in this category The mean, Let be the energy consumption factor of the i-th device. The energy consumption factor is obtained by comparing the historical energy consumption ratio data of different devices with the average energy consumption ratio data of devices of the same type. When the historical energy consumption ratio of a device is greater than the average data ratio, the energy consumption factor < 1; conversely, when the historical energy consumption ratio of a device is less than the average data ratio, the energy consumption factor > 1. Therefore, by adjusting the deviation amount through the energy consumption factor, the influence of the difference in energy consumption ratios of different devices on the device deviation comparison can be eliminated, thereby improving the accuracy of the judgment result. The deviation consistency coefficient s is compared with a preset threshold st. The preset threshold st is set based on the experience data of the corresponding type of device. Therefore, when s ≤ st, it indicates that the consistency of devices of the same type is high and the risk of abnormal power consumption is low. At this time, the reference power consumption of all devices is superimposed and compared with the total power consumption. If the comparison error exceeds the preset error threshold, it is judged that there is a risk of leakage; otherwise, it is judged that the device is operating normally. The preset error threshold is set with reference to the line loss data in the park. Therefore, when the comparison error exceeds the preset error threshold, it indicates that there is a high risk of leakage, and thus a leakage risk is judged. When s > st, it indicates that the power consumption status of some devices exceeds the average range, and therefore... The devices are sorted from largest to smallest. The data corresponding to the first device in the sort is excluded, and the deviation consistency coefficient s is recalculated. This process is repeated until s ≤ st. This allows us to identify devices that exceed the average range and classify them as risk devices, thus enabling the identification of devices with abnormal power consumption.
[0035] In one embodiment, a process for obtaining the start time point of energy consumption anomaly is provided, including: constructing an energy consumption anomaly start time point judgment model to obtain the energy consumption anomaly start time point tp, wherein the energy consumption anomaly start time point judgment model is represented as follows:
[0036] in, express The maximum value corresponds to the time point. A preset fixed time difference is selected based on the changes in abnormal power consumption curves in empirical data. In this embodiment, it is set to 20 seconds, which is chosen before the time point when the power consumption difference changes the most. Using the corresponding time point as the starting point of energy consumption anomalies can help to make more accurate analysis and judgment on the equipment status in the future.
[0037] In one embodiment, a process for assessing equipment risk is provided, including: determining the equipment's operating status after the start time of abnormal energy consumption based on the equipment's operating parameters; if the equipment is in a non-operating state but still consuming power, it is determined that the equipment has an abnormal activation problem; if the equipment is in an operating state, the temperature data of the equipment is obtained, a temperature analysis model is constructed according to different equipment types, and temperature anomaly factors are obtained. The temperature analysis model is expressed as follows:
[0038] Where T(t) is the equipment temperature change curve, and Te(t) is the ambient temperature change curve. For preset time periods; Based on the selected settings, the preset values are set by fitting the test data and satisfying the following conditions: >1, when the temperature anomaly factor If the power consumption exceeds the preset value, it indicates that the device is overheating abnormally, leading to increased power consumption, thus indicating a hardware malfunction. Otherwise, it indicates an abnormal device configuration. It should be noted that the main causes of abnormal power consumption include leakage, abnormal device startup, abnormal settings and configuration, and device malfunction. Device malfunction generally leads to an increase in device temperature. Therefore, the above comparison process can help determine the specific cause of abnormal power consumption, thus facilitating timely adaptive maintenance of the equipment by park management personnel.
[0039] In one embodiment, the management method of the smart park further includes: dynamically adjusting the patrol vehicle patrol strategy according to the first comparison result and the second comparison result, including: adjusting the value of the preset threshold st to sr, and satisfying sr < st; excluding data under the condition s ≤ sr as predicted risk devices; although the predicted risk devices have not triggered abnormal power consumption, their deviation exceeds the average level, so they have a high risk of abnormal power consumption; by obtaining the location of the predicted risk devices, increasing the frequency of patrol vehicle patrols at the location of the predicted risk devices, thereby ensuring that the devices can be dealt with in a timely manner when safety risks occur.
[0040] In one embodiment, a smart park management system based on an artificial intelligence big data model is provided, including an energy consumption data acquisition terminal, an equipment monitoring terminal, and a processor. The energy consumption data acquisition terminal is used to acquire environmental parameters and equipment operating parameters within the park and input them into the artificial intelligence big data model to obtain reference energy consumption data for each device. The equipment monitoring terminal is used to acquire monitoring data of the devices. The processor is used to perform a first comparison between the reference energy consumption data of each device and the real-time monitored energy consumption data to obtain a first comparison result, and to perform a second comparison between the first comparison results of devices of the same category to obtain a second comparison result. Based on the first comparison result and the second comparison result, it is determined whether there is an abnormality in the energy consumption status of the device. When it is determined that the energy consumption status of the device is abnormal, the starting time point of the energy consumption abnormality is obtained, and the risk of the device is judged based on the starting time point of the energy consumption abnormality and the monitoring data of the device. By comparing the reference energy consumption data of each device with the real-time monitored energy consumption data, abnormal power consumption of the equipment can be judged more accurately and timely. By comparing the results of the first comparison with the first comparison of similar equipment, the abnormal power consumption of the equipment can be further judged, thereby improving the accuracy and timeliness of the judgment of abnormal power consumption. By obtaining the starting time point of the abnormal power consumption, the risk of the equipment can be judged based on the starting time point of the abnormal power consumption and the monitoring data of the equipment. This allows for timely auxiliary judgment of the cause of the abnormal power consumption, which facilitates the park management personnel to maintain the equipment in a timely manner based on the judgment results, ensuring the stability and energy saving of the equipment operation.
[0041] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.
Claims
1. A smart park management method based on a large-scale artificial intelligence model, characterized in that, The method comprises: The environmental parameters and equipment operating parameters within the park are acquired and input into the large-scale artificial intelligence model to obtain reference energy consumption data for each device; The reference energy consumption data of each device is compared with the real-time monitored energy consumption data to obtain the first comparison result. The first comparison results of the same type of devices are then compared to obtain the second comparison result. Based on the first comparison result and the second comparison result, it is determined whether there is any abnormality in the energy consumption status of the device. When an abnormal energy consumption status of a device is detected, the starting time of the abnormal energy consumption is obtained, and the risk of the device is assessed based on the starting time of the abnormal energy consumption and the device's monitoring data.
2. The smart park management method based on an artificial intelligence large-scale model according to claim 1, characterized in that, The process of obtaining reference energy consumption data for each device includes: The environmental parameters and operating parameters of each device within an analysis period are obtained and input into the large artificial intelligence model to obtain the reference power consumption of the device within the analysis period. The reference power consumption of the continuous analysis cycle is placed in a plane coordinate system, and the points in the coordinate system are fitted to obtain the reference power consumption curve of each device. The reference power consumption curve is used as the reference energy consumption data.
3. The smart park management method based on an artificial intelligence large-scale model according to claim 2, characterized in that, The first comparison process includes: The real-time monitored energy consumption data is fitted to a real-time power consumption curve E(t). The real-time power consumption curve E(t) and the reference power consumption curve Er(t) are placed in the same coordinate system, and a comparison model is constructed to obtain the deviation dev. The comparison model is expressed as follows: ； Where tx is the start time point and ty is the end time point; Compare the deviation amount dev with the preset error tolerance: When the deviation dev exceeds the preset error allowable amount, it is determined that there is an abnormality in the energy consumption status; A second comparison is performed when the deviation dev is less than or equal to the preset error allowable amount.
4. The smart park management method based on an artificial intelligence large-scale model according to claim 3, characterized in that, The second comparison process includes: An energy consumption dispersion model is constructed for each device, and the deviation consistency coefficient s of each device is obtained. The energy consumption dispersion model is represented as follows: ； Where n is the number of devices in this category, and i is a positive integer and i∈[1,n]. Let be the deviation of the i-th device. Let i be the energy consumption factor of the i-th device. For all devices in this category The mean; Compare the deviation consistency coefficient s with the preset threshold st: If s≤st, the reference power consumption of all devices is summed and compared with the total power consumption. If the comparison error exceeds the preset error threshold, it is determined that there is a risk of leakage; otherwise, it is determined that the device is operating normally. If s > st, according to Sort the devices from largest to smallest, exclude the data corresponding to the first device in the sort, recalculate the deviation consistency coefficient s, and repeat this process until s≤st. The devices whose data has been excluded are then identified as risk devices.
5. The smart park management method based on an artificial intelligence large-scale model according to claim 4, characterized in that, The process of obtaining the start time point of energy consumption anomalies includes: A model for determining the starting time point of energy consumption anomalies is constructed to obtain the starting time point tp of energy consumption anomalies. The model for determining the starting time point of energy consumption anomalies is represented as follows: ； in, express The maximum value corresponds to the time point. Preset fixed time difference.
6. The smart park management method based on an artificial intelligence large-scale model according to claim 5, characterized in that, The process of assessing equipment risk includes: Determine the equipment's operating status after the initial time point of abnormal energy consumption based on the equipment's operating parameters: If the device is not in operation, it is determined that the device has been abnormally activated; If the equipment is in operation, acquire its temperature data, construct a temperature analysis model based on different equipment types, and identify temperature anomaly factors. The temperature analysis model is expressed as follows: ； Where T(t) is the equipment temperature change curve, and Te(t) is the ambient temperature change curve. For preset time periods; When temperature anomaly factor If the value exceeds the preset value, the device is considered to have a hardware malfunction; otherwise, the device is considered to have an abnormal configuration.
7. The smart park management method based on an artificial intelligence large-scale model according to claim 6, characterized in that, The method further includes: The patrol vehicle patrol strategy is dynamically adjusted based on the first and second comparison results, including: Adjust the value of the preset threshold st to sr, and satisfy sr < st. Select the devices that exclude data under the condition s ≤ sr as the predicted risk devices, obtain the location points of the predicted risk devices, and increase the frequency of patrol vehicles patrolling the location points of the predicted risk devices.
8. A smart park management system based on a large-scale artificial intelligence model, characterized in that, The system is used to execute a smart park management method based on an artificial intelligence large model as described in any one of claims 1-7, including an energy consumption data acquisition terminal, an equipment monitoring terminal, and a processor; The energy consumption data acquisition terminal is used to acquire environmental parameters and equipment operating parameters within the park and input them into the artificial intelligence big model to obtain reference energy consumption data for each device; The device monitoring terminal is used to acquire monitoring data from the device. The processor is used to perform a first comparison between the reference energy consumption data of each device and the real-time monitored energy consumption data to obtain a first comparison result, perform a second comparison between the first comparison results of devices of the same category to obtain a second comparison result, and determine whether there is an abnormality in the energy consumption status of the device based on the first comparison result and the second comparison result. When an abnormal energy consumption status of a device is detected, the starting time of the abnormal energy consumption is obtained, and the risk of the device is assessed based on the starting time of the abnormal energy consumption and the device's monitoring data.