Hotel energy consumption monitoring management system based on big data

Abstract

The invention discloses a hotel energy consumption monitoring management system based on big data, and relates to the technical field of energy consumption monitoring management, low-efficiency equipment is screened out according to equipment energy consumption data and operation feedback data, and operation parameters of the low-efficiency equipment are automatically controlled by using a trained fuzzy control model by taking energy consumption reduction as a control target; a pre-trained reinforcement learning algorithm is used to optimize a selected operation mode, feedback data after optimization is collected, an improvement degree is constructed according to changes of energy consumption data, and if the obtained improvement degree does not exceed a preset improvement threshold value, fault detection is carried out on equipment, and alarm information is sent out when a fault exists. And outputting a corresponding energy consumption optimization scheme by the hotel equipment energy consumption maintenance knowledge graph, and constraining a maintenance period. The constrained maintenance period constraint can prevent hotel equipment from being in a long-term sub-health state, repeated or delayed maintenance is avoided, and the availability rate of the equipment is improved while energy consumption is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of energy consumption monitoring and management, and in particular to a hotel energy consumption monitoring and management system based on big data. Background Art

[0002] With the rapid development of the global tourism industry, as an energy-intensive industry, the energy consumption management and optimization of hotels are of great significance in improving operational efficiency, reducing operating costs and achieving sustainable development. Modern hotels cover multiple functional areas, including guest rooms, public areas and logistics areas, which are equipped with a large number of high-energy-consuming equipment, such as central air conditioners, heat pumps, boilers, lighting systems and various household appliances. The efficient operation of these equipment not only directly affects the energy cost of the hotel, but also has a significant impact on the comfort and overall experience of the guests. Therefore, how to achieve real-time energy consumption monitoring, accurate energy consumption analysis and intelligent control of various areas and equipment of the hotel has become a key issue in improving the management level and competitiveness of the hotel. To this end, existing technologies have attempted to achieve some energy consumption management goals by deploying intelligent sensors, energy consumption monitoring systems and basic data analysis tools. However, these existing systems still have many shortcomings in practical applications.

[0003] On the other hand, with the rapid development of the Internet of Things (IoT), big data analysis and artificial intelligence (AI) technologies, how to apply these advanced technologies to hotel energy management systems has become the key to solving the above technical problems. Big data technology can process and analyze massive data from various sensors and devices, and identify energy consumption patterns and potential optimization opportunities through data mining and machine learning algorithms. Artificial intelligence technology, especially deep learning and reinforcement learning, can further improve the intelligence level of the system, achieve accurate prediction and dynamic control of energy consumption, optimize equipment operation strategies, and improve overall energy efficiency. However, most of the hotel energy management systems on the market currently lack deep integration of big data technology, and cannot make full use of massive data resources for intelligent analysis and decision-making, resulting in unsatisfactory results in energy optimization and equipment maintenance.

[0004] In the Chinese invention patent with application publication number CN111652457A, a method, system, electronic device and storage medium for managing energy consumption of a smart hotel are disclosed, the method comprising: obtaining a first unit area energy consumption and a first unit room energy consumption, the first unit area energy consumption is the average of the unit area energy consumption of all hotels, and the first unit room energy consumption is the average of the unit room energy consumption of all hotels; obtaining a second unit area energy consumption and a second unit room energy consumption, the second unit area energy consumption is the unit area energy consumption of the hotel to be managed, and the second unit room energy consumption is the unit room energy consumption of the hotel to be managed; according to the first unit area energy consumption, the second unit area energy consumption, the first unit room energy consumption and the second unit room energy consumption, the energy consumption of the hotel to be managed is controlled. The corresponding system, electronic device and storage medium also disclosed in the present invention can effectively control the energy consumption of the hotel to be managed.

[0005] Combining the above applications and the prior art: Traditional energy consumption monitoring systems mostly rely on manual recording or a single type of automated equipment, which makes it difficult to fully cover all high-energy consumption equipment in the hotel, resulting in incomplete or delayed energy consumption data collection, and unable to provide real-time and accurate energy consumption information. In addition, the energy consumption analysis methods of existing systems are mostly based on static rules or basic statistical models, lacking the ability to dynamically respond to complex and changeable external environmental factors and internal equipment status. For example, weather changes such as temperature, humidity, wind speed, etc. have a direct impact on the energy consumption of the air-conditioning system. At the same time, the existing system has a low degree of intelligence in equipment fault detection and preventive maintenance, and usually relies on post-maintenance or basic fault alarms, making it difficult to achieve early warning and intelligent maintenance decisions. These defects not only limit the effectiveness of the energy consumption management system, but also increase energy waste and maintenance costs, hindering the hotel's comprehensive transformation towards intelligence and greenness.

[0006] The present invention provides a hotel energy consumption monitoring and management system based on big data. Summary of the invention

[0007] 1. Technical issues to be resolved In view of the deficiencies of the prior art, the present invention provides a hotel energy consumption monitoring and management system based on big data. By screening out inefficient equipment based on equipment energy consumption data and operation feedback data, reducing energy consumption as the control goal, and using the trained fuzzy control model to automatically control the operating parameters of inefficient equipment; using the pre-trained reinforcement learning algorithm to optimize the selected operation mode, collecting feedback data after execution optimization, and constructing the improvement degree according to the change of energy consumption data, if the obtained improvement degree does not exceed the preset improvement threshold, the equipment is detected for faults and an alarm message is issued when a fault exists, and the hotel equipment energy consumption maintenance knowledge map outputs the corresponding energy consumption optimization plan, and constrains the maintenance cycle. Prevent hotel equipment from being in a long-term sub-healthy state, avoid repetitive or delayed maintenance, and improve the availability of equipment while reducing energy consumption; thereby solving the technical problems raised in the background technology.

[0008] (II) Technical solution To achieve the above objectives, the present invention is implemented through the following technical solutions: a hotel energy consumption monitoring and management system based on big data, including a correlation analysis unit, which collects equipment status and environmental condition data of each area of ​​the hotel. If the weather conditions are abnormal, the energy consumption data and weather scores in the sub-area are obtained and correlation analysis is performed, and the influence degree is constructed from the correlation data. , if the influence Exceeding expectations and issuing analytical instructions to external parties; The equipment optimization control unit collects the energy consumption data of each device for visualization, selects inefficient equipment based on the equipment energy consumption data and operation feedback data, takes reducing energy consumption as the control target, and uses the trained fuzzy control model to automatically control the operating parameters of inefficient equipment; The adjustment feedback unit collects the user's check-in status data, matches the user with the corresponding operation mode, collects the user's active adjustment data on the equipment in real time as feedback, optimizes the selected operation mode using the pre-trained reinforcement learning algorithm, and adds the optimized operation mode to the operation mode library; The energy consumption improvement analysis unit collects feedback data after optimization and builds improvement measures based on changes in energy consumption data. , if the improvement obtained is If the preset improvement threshold is not exceeded, a detection instruction is issued to the outside; The maintenance unit uses the operating status data of the inefficient equipment as input, detects equipment faults and issues alarm information when a fault occurs. The alarm density in the hotel area is calculated based on the alarm related information. , if the alarm density obtained If it exceeds expectations, the hotel equipment energy consumption maintenance knowledge graph will output the corresponding energy consumption optimization plan and constrain the maintenance cycle.

[0009] Furthermore, smart sensors are installed in various areas of the hotel, and energy consumption monitoring modules are installed for high-energy consumption equipment; equipment operating status data and environmental condition data are collected in real time; the collected weather data is used as input, and the trained weather anomaly recognition model is used to perform anomaly recognition and determine the degree of weather anomaly. If an anomaly exists, a weather score evaluating the current weather anomaly is output.

[0010] Furthermore, after constructing an electronic map covering the hotel area, the hotel area is divided into several sub-areas, and the sub-areas are marked with correlation data and influence indexes are constructed. , as follows: ; Where: , , i is the sub-region number, ; To calculate the obtained i The closeness centrality of each sub-region is For the i The correlation coefficient of the sub-regions, is the corresponding mean.

[0011] Furthermore, after receiving the analysis instruction, the energy consumption data of each device is recorded in real time, classified by region or device type, and the power consumption and usage time of the device are recorded; The collected equipment energy consumption data and operation feedback data are used as input, and the trained energy consumption identification model is used to evaluate the equipment energy consumption. If the corresponding evaluation result does not exceed the expectation, the corresponding equipment will be marked as inefficient equipment; the location information of the inefficient equipment is obtained, and the operating parameters of the inefficient equipment are collected.

[0012] Furthermore, a smart platform is established in the check-in area, the lighting system, air conditioning, curtains and other equipment of the guest rooms are connected to the smart platform, and a dynamic sensing network is deployed to obtain the user's check-in status data and collect the user's check-in preference data; Several operating modes are designed for the equipment in the occupancy area, and an operating mode library is generated after being summarized; the user's occupancy preference data, external weather condition data, and occupancy status data are used as input, and a content-based recommendation algorithm matches the user with the corresponding operating mode from the operating mode library.

[0013] Furthermore, the user's check-in feedback data is collected within a preset feedback cycle, and the feedback data is used as input again to automatically optimize the control parameters of the current operation mode using a pre-trained Bayesian update algorithm to obtain an optimized mode; Collect the energy consumption data of equipment in the optimized occupied area within the preset observation period, and build improvement measures based on the changes in energy consumption data. , as follows: ; Where: is the starting time point of the observation period, T is the end time of the observation period, For time t The baseline energy consumption at the time, For time t Optimized energy consumption at all times, is the rate of change of energy consumption over time; is the time decay parameter, , is the weight coefficient, .

[0014] Furthermore, the operating status data of various inefficient devices are collected in real time, and the equipment operating data is used as input. The trained fault detection model is used for fault detection. If the equipment operating status is abnormal, it is treated as an abnormal device and an alarm instruction is issued to the outside. When the number of consecutive alarm instructions received in the current stage exceeds the expectation, the relevant alarm information is summarized, and the alarm density in the hotel area is calculated based on the alarm-related information. ; If the alarm density obtained exceeds expectations, the abnormal equipment will be marked on the electronic map, and the pre-trained path planning will plan a maintenance path for the abnormal equipment based on the location information and degree of abnormality of the abnormal equipment.

[0015] Furthermore, the alarm density in the hotel area is calculated based on the alarm-related information. The way is as follows: ; Where: A is the hotel area, T To observe the cycle length, is the starting time point of the observation period; For the hotel's space domain, For the location x and time t The degree of abnormality of the equipment at the location; For the location x and time t Weather condition ratings for is the adjustment coefficient, , is the time decay parameter.

[0016] Furthermore, feature extraction is performed after collecting detection data and energy consumption data of abnormal equipment to obtain corresponding maintenance features; abnormal energy consumption maintenance of hotel electrical equipment is used as the target word, and a knowledge graph of hotel equipment energy consumption maintenance is pre-constructed; According to the correspondence between maintenance characteristics and energy consumption optimization plans, the hotel equipment energy consumption maintenance knowledge map outputs the corresponding energy consumption optimization plan, executes the energy consumption optimization plan to maintain abnormal equipment and constrains the maintenance interval .

[0017] Furthermore, after a maintenance interval that meets the constraints After that, maintenance constraints are applied to abnormal equipment. The constraint methods are as follows: ; Where: For maintenance intervals, T is the current time point, t is the time variable, For time t The device abnormality score at the moment, For time t Weather condition rating at the moment, is the weather condition influence coefficient, ; is the time decay parameter, ranging from 0 to 1. is the cumulative abnormality threshold. When this threshold is reached or exceeded, maintenance is triggered. Minimum maintenance intervals to ensure that maintenance operations are not too frequent. .

[0018] (III) Beneficial effects The present invention provides a hotel energy consumption monitoring and management system based on big data, which has the following beneficial effects:

[0019] 1. When the correlation results show that the energy consumption of a certain area is abnormally obvious under extreme weather conditions, dynamic monitoring is formed. As the weather and occupancy conditions change, the regional energy consumption risk assessment is continuously updated, achieving a close coupling between meteorological information and the hotel energy system.

[0020] 2. By training the energy consumption identification model to accurately evaluate the energy consumption performance of each device, and combining the fuzzy control algorithm to implement automatic control of inefficient equipment, it can significantly reduce energy consumption and improve hotel operation and maintenance efficiency. The identified inefficient equipment can be located and its parameters tracked, and the equipment can be optimized or replaced in a targeted manner, which can effectively reduce the redundant operation of high-energy-consuming equipment and significantly reduce unnecessary energy waste.

[0021] 3. By deploying an intelligent control platform in the check-in area and supplemented by a dynamic sensing network, the hotel can achieve refined energy consumption control of guest rooms and public areas. The collection of guest preference data can provide key basis for subsequent algorithm recommendations and pattern matching, while providing personalized settings for guests and dynamically maintaining a reasonable range of energy consumption.

[0022] 4. By combining user preference data with external environmental information, the system can accurately match the mode for guests, quickly respond to scene changes and make corresponding energy consumption scheduling; by combining user preferences with external weather data, the system can quickly complete mode switching when the user just enters the room or as the weather changes, effectively reducing energy consumption.

[0023] 5. Automatic optimization of the default parameters of the operating mode can enhance the sophistication and adaptability of the hotel's control strategy; it can effectively reduce energy waste and improve guest satisfaction; it can detect whether there are abnormal conditions in the equipment in real time, and trigger alarms in time when potential equipment failures are discovered to prevent the failures from expanding and causing higher energy waste or safety risks, thus transforming equipment maintenance from passive inspection to active early warning.

[0024] 6. Quickly check and locate areas or equipment groups that may have high-risk hidden dangers, shortening the time for locating faults; combining weather information, time information, etc. to help determine whether it is related to extreme weather or special usage scenarios; it can help maintenance personnel efficiently complete inspections and troubleshooting in large or multi-story buildings, thereby improving maintenance efficiency.

[0025] 7. Using the constructed hotel equipment energy consumption maintenance knowledge graph to match the corresponding energy consumption optimization plan can significantly improve the accuracy and efficiency of maintenance decisions. The constrained maintenance cycle constraints can prevent hotel equipment from being in a sub-healthy state for a long time, avoid repetitive or delayed maintenance, and reduce energy consumption while improving the availability and service life of equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 This is a flow chart of a hotel energy consumption monitoring and management method based on big data according to the present invention; Figure 2 It is a structural schematic diagram of the hotel energy consumption monitoring and management system based on big data of the present invention. DETAILED DESCRIPTION

[0027] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] Traditional energy consumption monitoring systems mostly rely on manual recording or a single type of automated equipment, which makes it difficult to fully cover all high-energy consumption equipment in the hotel, resulting in incomplete or delayed energy consumption data collection, and unable to provide real-time and accurate energy consumption information. In addition, the energy consumption analysis methods of existing systems are mostly based on static rules or basic statistical models, lacking the ability to dynamically respond to complex and changeable external environmental factors and internal equipment status. For example, weather changes such as temperature, humidity, wind speed, etc. have a direct impact on the energy consumption of the air-conditioning system. At the same time, the existing system has a low degree of intelligence in equipment fault detection and preventive maintenance, and usually relies on post-maintenance or basic fault alarms, making it difficult to achieve early warning and intelligent maintenance decisions. These defects not only limit the effectiveness of the energy consumption management system, but also increase energy waste and maintenance costs, hindering the hotel's comprehensive transformation towards intelligence and greenness.

[0029] See also Figure 1 The present invention provides a hotel energy consumption monitoring and management method based on big data, including: Step 1: After collecting the equipment status and environmental condition data of each area of ​​the hotel, if the weather conditions are abnormal, obtain the energy consumption data and weather scores in the sub-area and perform correlation analysis, and construct the impact degree from the correlation data. , if the influence Exceeding expectations and issuing analytical instructions to external parties; The step 1 includes the following contents: Step 101, install smart sensors such as energy consumption monitoring meters, power grid meters, temperature and humidity monitors, etc. in various areas of the hotel, such as guest rooms, public areas and logistics areas; install independent energy consumption monitoring modules for high-energy consumption equipment, such as central air conditioners, heat pumps and boilers; collect equipment operation status data in guest rooms and public areas in real time, such as air conditioning temperature and lighting status, and environmental condition data, such as temperature and humidity, external temperature and other weather conditions; pre-process the collected data, and upload the sensor data to the cloud platform in real time; Specifically, by installing data acquisition modules inside or outside of air conditioners, lighting and other equipment, the operating hours, on / off status and power consumption are recorded in real time, and the data is uploaded to the cloud platform for aggregation via wireless or wired networks, ensuring that the energy consumption of each area can be monitored and analyzed in real time on a unified platform; Through the above method, multiple types of smart sensors are installed in various areas of the hotel, and data pre-processing and real-time uploading are performed to improve the accuracy and timeliness of hotel energy consumption data collection. Monitoring modules are independently installed on air conditioners, lighting and other equipment to accurately quantify the energy consumption of each type of equipment, allowing managers to quickly locate high energy consumption sources and perform targeted optimization. Step 102: train a convolutional neural network with the labeled sample data to obtain a trained weather anomaly recognition model; collect external weather data in the current stage, such as temperature, humidity, wind speed, weather type, etc., and aggregate them to generate a weather data set; use the collected weather data as input, use the trained weather anomaly recognition model to perform anomaly recognition, and judge the degree of weather anomaly. If there is an anomaly, such as when the external environment temperature deviation exceeds a certain range or the weather conditions are extreme, output a weather score that evaluates the current weather anomaly; When in use, the "weather score" is calculated in real time during runtime, and preventive or responsive energy consumption configuration plans can be made in a timely manner according to external extreme weather conditions (such as extreme cold, extreme heat, strong winds and rainstorms, etc.), maintaining high sensitivity detection of environmental changes, and helping the hotel maintain stable and energy-saving operation under sudden weather conditions.

[0030] Step 103: After constructing the electronic map covering the hotel area, the hotel area is divided into a plurality of sub-areas; After obtaining the energy consumption data and weather scores in the sub-areas, correlation analysis is performed to obtain the correlation data. Under dimensionless conditions, the sub-areas are marked with the correlation data and the influence degree is constructed. , as follows: ; Where: , , i is the sub-region number, ; To calculate the obtained i The closeness centrality of each sub-region is For the i The correlation coefficient of the sub-regions, is the corresponding mean; If the obtained impact exceeds the preset impact threshold, it means that the current weather conditions have a greater impact on the energy consumption of equipment in the hotel area. At this time, an analysis instruction is issued to the outside; When using, combine the contents in steps 101 to 103: By constructing a visual electronic map within the hotel and dividing the overall area into several sub-areas, the accuracy and timeliness of the hotel's energy consumption management can be refined; once the energy consumption of a sub-area suddenly increases, the extent to which the sub-area may be affected by external extreme weather or its own equipment abnormalities can be quickly located. When the correlation results show that the energy consumption of a certain area is obviously abnormal under extreme weather, this area can be given priority for in-depth investigation or optimization, thereby effectively improving the energy efficiency of the hotel's overall operations; dynamic monitoring is formed, and the regional energy consumption risk assessment is continuously updated as the weather and occupancy conditions change, realizing the close coupling between meteorological information and the hotel's energy system.

[0031] Closeness centrality is a metric used to measure the average distance between a node and all other nodes in a network. In the hotel energy consumption management system, closeness centrality can be used to evaluate the importance of each sub-area or device in the entire energy consumption network. Specifically, sub-areas or devices with higher closeness centrality can transmit and respond to information with other parts of the network more quickly, thus playing a key role in energy consumption monitoring, fault diagnosis and optimization control. By analyzing the closeness centrality of each sub-area, managers can identify key nodes that are at the core of the energy consumption management system, and prioritize the optimization and resource allocation of these high-impact areas. This not only helps to improve the overall energy consumption efficiency, but also enhances the response speed and stability of the system, and realizes more intelligent and efficient hotel operation management.

[0032] Step 2: Collect energy consumption data of each device for visualization, select inefficient devices based on the device energy consumption data and operation feedback data, take reducing energy consumption as the control goal, and use the trained fuzzy control model to automatically control the operating parameters of inefficient devices; The step 2 includes the following contents: Step 201: After receiving the analysis instruction, the power consumption, operation time, peak power and other energy consumption data of each device are recorded in real time, and classified by area (guest room, public area, logistics area, etc.) or equipment type (lighting, HVAC, water pump, etc.); After recording the power consumption and usage time of each device, use BI reports or customized web interfaces to present real-time energy consumption curves, regional energy consumption distribution, equipment energy consumption rankings, etc. in the form of charts, and support comparison with historical data for the same period, so as to facilitate identification of abnormally high energy consumption periods or areas, and visualize energy consumption data in each sub-area; When in use, the equipment energy consumption data is collected and classified and archived in real time, and displayed in combination with visualization tools, which can quickly provide insights into the distribution of energy consumption and a more intuitive understanding of the energy consumption patterns in different time periods, different regions or different equipment types, thereby continuously reducing energy expenditures and improving overall operational efficiency in the long term.

[0033] Step 202: Use the labeled sample data to train the machine learning algorithm to obtain a trained energy consumption identification model; use the collected equipment energy consumption data and operation feedback data as input, use the trained energy consumption identification model to evaluate the equipment energy consumption, and if the corresponding evaluation result does not exceed the expectation, mark the corresponding equipment as an inefficient equipment; obtain the location information of the inefficient equipment, and collect the operation parameters of the inefficient equipment; The fuzzy control algorithm is trained by the labeled sample data to obtain the trained fuzzy control model. The trained fuzzy control model is used to automatically control the operating parameters of the inefficient equipment with the reduction of energy consumption as the control target. When using, combine the contents in steps 201 and 202: By training the energy consumption identification model to accurately evaluate the energy consumption performance of each device, and combining the fuzzy control algorithm to implement automated control of inefficient equipment, it can significantly reduce energy consumption and improve hotel operation and maintenance efficiency, avoiding the shortcomings of relying solely on experience or rough indicators for judgment; locating and tracking parameters of identified inefficient equipment, optimizing or replacing equipment in a targeted manner, and reducing unnecessary energy losses; balancing temperature, load, and user comfort to make equipment adjustments more in line with actual application needs, and accurate and adaptive control methods can effectively reduce the redundant operation of high-energy-consuming equipment and significantly reduce unnecessary energy waste.

[0034] Step 3: After collecting the user's check-in status data, match the user with the corresponding operation mode, collect the user's active adjustment data of the equipment in real time as feedback, use the pre-trained reinforcement learning algorithm to optimize the selected operation mode, and add the optimized operation mode to the operation mode library; The step three includes the following contents: Step 301: Establish an intelligent control platform in the check-in area, connect the lighting system, air conditioning, curtains and other equipment of the guest room to the intelligent control platform, and also equip it with a voice assistant or a smart panel; deploy a dynamic sensing network, such as setting up infrared sensors or cameras, etc., to detect whether there is someone in the room, whether they are sleeping, etc., and obtain the user's check-in status data through the dynamic sensing network; In addition, people counting devices can be installed in corridors and meeting rooms to automatically adjust lighting brightness and air conditioning speed according to real-time traffic flow to avoid energy waste. At this time, the user's stay preference data is collected, for example, preferences for environmental conditions in the stay area, including temperature, humidity and wind speed; By deploying an intelligent control platform in the check-in area and supplemented by a dynamic sensing network, the hotel can achieve refined energy consumption control of guest rooms and public areas, while greatly improving the comfort experience of guests; unified access to lighting, air conditioning, curtains, etc. to achieve integrated management and reduce repeated energy consumption; through accurate perception of personnel behavior, it saves energy without affecting user use; collecting guest preference data can provide key basis for subsequent algorithm recommendations and pattern matching, so that the system can dynamically maintain a reasonable range of energy consumption while providing guests with a more intimate comfort experience and personalized settings.

[0035] Step 302: Design several operation modes for the equipment in the occupancy area, such as occupancy mode, departure mode, eco-friendly mode and sleep mode, and generate an operation mode library after summarizing; use the user's occupancy preference data, external weather condition data and occupancy status data as input, and use the content-based recommendation algorithm to match the user with the corresponding operation mode from the operation mode library; for example, automatically enter the low energy consumption mode when the guest leaves the room or at night; By designing multiple operating modes for the occupancy area and building an operating mode library, combining user preference data with external environmental information, and using content-based recommendation algorithms to accurately match modes for guests, the hotel's intelligent and customized service level can be greatly improved, and clear logic and configuration are provided for the automatic operation of the equipment, so that the system can quickly respond to scene changes and perform corresponding energy consumption scheduling; combining user preferences with external weather data, the mode switch can be quickly completed when the user just enters the room or as the weather changes, ensuring that the guest is always in the most comfortable environment; and when the user leaves the room or when night time comes, it can automatically enter the low-energy mode, effectively reducing energy consumption.

[0036] Step 303: collect the user's active adjustment data on electrical equipment such as temperature and lighting in real time, use the active adjustment data as feedback, use the pre-trained reinforcement learning algorithm to optimize the selected operation mode, continuously improve the balance between comfort and energy saving effect, and add the optimized operation mode to the operation mode library; When used, combine the contents in steps 301 and 302: By real-time monitoring of users' active adjustment behaviors and iterative optimization of operating modes using reinforcement learning algorithms, the optimal balance between comfort and energy consumption can be continuously found in a dynamic occupancy environment. Reinforcement learning will automatically update mode parameters based on users' frequent operation of the equipment or fine-tuning data, so that the entire system can gradually find the best configuration that takes into account both guest experience and energy savings during operation.

[0037] Step 4: Collect feedback data after optimization and build improvement data based on changes in energy consumption data , if the improvement obtained is If the preset improvement threshold is not exceeded, a detection instruction is issued to the outside; The step 4 includes the following contents: Step 401: Taking a period of the expected stay of the user as a feedback cycle, collecting the user's stay feedback data within the preset feedback cycle, for example, counting the frequency and magnitude of the user's manual changes in air-conditioning temperature, light brightness, curtain position, etc.; if the user frequently makes manual adjustments in a certain operation mode, it indicates that the current default settings are not appropriate; Using the feedback data as input again, the pre-trained Bayesian update algorithm is used to automatically optimize the control parameters of the current operating mode, optimize each operating mode, and obtain the optimized mode; Automatic optimization of the default parameters of the operating mode can enhance the sophistication and adaptability of the hotel's control strategy; the use of the Bayesian update algorithm can quickly correct the existing mode parameters when new data arrives, which can ensure the stability and objectivity of the mode optimization, effectively reduce energy waste and improve guest satisfaction.

[0038] Step 402: Collect the energy consumption data of the equipment in the optimized occupied area within the preset observation period, and construct the improvement degree according to the change of energy consumption data under dimensionless conditions. , to determine the current energy consumption optimization effect, the method is as follows: ; Where: is the starting time point of the observation period, T is the end time of the observation period, For time t The baseline energy consumption at the time, For time t Optimized energy consumption at all times, is the rate of change of energy consumption over time; is the time decay parameter, with a value between 0 and 1. , is the weight coefficient, ; e The value can be 2.718; Based on historical data and management expectations for energy consumption optimization, an improvement threshold is pre-set. If the improvement degree obtained does not exceed the preset improvement threshold, the improvement effect does not meet expectations, indicating that some of the current equipment may be difficult to effectively control or optimize, and the energy consumption is still high. At this time, a detection instruction is issued to the outside; and the real-time energy consumption data of guest rooms, public areas and equipment is visualized, and detailed energy consumption reports are generated in stages; When using, combine the contents in steps 401 and 402: If the improvement Higher, indicating that the current optimization path is relatively successful; if the improvement If the value is below the threshold, it can quickly attract attention, trigger external detection instructions, and take other supplementary or alternative measures in time; real-time visual display can help operation and maintenance personnel quickly focus on specific areas or equipment, shorten the troubleshooting cycle or optimize the adjustment cycle.

[0039] Step 5: Using the operating status data of the inefficient equipment as input, perform fault detection on the equipment and send out alarm information when a fault occurs. The alarm density in the hotel area is calculated based on the alarm related information. , if the alarm density If it exceeds expectations, the hotel equipment energy consumption maintenance knowledge graph will output the corresponding energy consumption optimization plan and constrain the maintenance cycle; The step five includes the following contents: Step 501: collect the operating status data of various inefficient devices in real time, including operating time, current and voltage, energy consumption data and various control parameters, and summarize them as device operating data; The machine learning algorithm is trained with the labeled sample data to obtain the trained fault detection model. The equipment operation data is used as input and the trained fault detection model is used to perform fault detection. If the equipment operation status is abnormal, it is treated as an abnormal device and an alarm instruction is issued to the outside. With the help of trained fault detection models, we can detect in real time whether there are abnormal conditions in the equipment. When potential equipment faults are found, alarms are triggered in time to prevent the faults from further expanding and causing higher energy waste or safety risks. Intelligent fault warning can reduce the frequency of personnel inspections and transform equipment maintenance from passive maintenance to active warning. Step 502: When the number of consecutive alarm instructions received in the current stage exceeds the expected number, relevant alarm information is summarized, including the alarm location information, time information, weather condition information, and equipment abnormality degree, etc. Under dimensionless conditions, the alarm density in the hotel area is calculated based on the alarm related information. , as follows: ; Where: A is the hotel area, T To observe the cycle length, is the starting time point of the observation period; For the hotel's space domain, For the location x and time t The degree of abnormality of the equipment at the location; For the location x and time t Weather condition ratings for is the adjustment coefficient, , is the time decay parameter, ranging from 0 to 1; Pre-set intensive thresholds based on historical data and management expectations for equipment abnormal alarms; If the alarm density obtained exceeds the preset density threshold, it means that the equipment in the hotel area is operating poorly. The abnormal equipment will be marked on the electronic map. Based on the location information and abnormality level of the abnormal equipment, the pre-trained path planning will plan a maintenance path for the abnormal equipment. When in use, the introduction of the alarm density indicator can help hotel managers quickly check and locate areas or equipment groups that may have high-risk hidden dangers, facilitate rapid assessment of whether abnormalities are concentrated, whether they have spread to multiple areas, etc.; make obvious marks on the visual electronic map to shorten the time for locating faults; at the same time, combined with weather information, time information, etc., it can also help determine whether it is related to extreme weather or special usage scenarios; formulate the optimal maintenance path for areas and equipment with frequent alarms, which can help maintenance personnel efficiently complete inspections and troubleshooting in large or multi-story buildings, improve maintenance efficiency, and effectively prevent local out-of-control of equipment abnormal events, ensure that the normal operation of the equipment is restored in the shortest time and reduce further energy waste or potential safety risks.

[0040] Step 503: After collecting the detection data and energy consumption data of abnormal equipment, feature extraction is performed to obtain corresponding maintenance features; abnormal energy consumption maintenance of hotel electrical equipment is used as the target word, and after deep retrieval and entity relationship building, a knowledge graph of hotel equipment energy consumption maintenance is pre-built; According to the correspondence between maintenance characteristics and energy consumption optimization plans, the hotel equipment energy consumption maintenance knowledge map outputs the corresponding energy consumption optimization plan, executes the energy consumption optimization plan to maintain abnormal equipment and constrains the maintenance interval , perform maintenance on abnormal equipment after a maintenance interval that meets the constraints. The constraints are as follows: ; Where: For maintenance intervals, T is the current time point, t is the time variable, For time t The device abnormality score at the moment, For time t Weather condition rating at the moment, is the weather condition influence coefficient, ; is the time decay parameter, ranging from 0 to 1. is the cumulative abnormality threshold. When this threshold is reached or exceeded, maintenance is triggered. Minimum maintenance intervals to ensure that maintenance operations are not too frequent. ; When using, combine the contents in steps 501 to 503: By extracting multi-dimensional features from the detection data and energy consumption data of abnormal equipment, and then using the constructed hotel equipment energy consumption maintenance knowledge graph to match the corresponding energy consumption optimization plan, the accuracy and efficiency of maintenance decisions can be significantly improved. The constrained maintenance cycle constraints can prevent hotel equipment from being in a long-term sub-healthy state and avoid repetitive or delayed maintenance. While reducing energy consumption, it also increases the availability and service life of equipment; allowing the hotel to always maintain an efficient and low-consumption operating level between daily operations and sudden failures.

[0041] The process of building a knowledge graph for hotel equipment energy consumption maintenance includes the following key steps: First, collect multi-source heterogeneous data, including equipment archives, real-time energy consumption monitoring data, historical maintenance records, fault logs, manufacturer technical manuals and relevant industry standards, to ensure comprehensive coverage of all types of equipment information.

[0042] Secondly, data extraction, cleaning and standardization are carried out. Data preprocessing technology is used to remove noise, fill missing values ​​and unify data formats, so that data from different sources can be consistent at the semantic level.

[0043] Next, entity recognition and relationship extraction algorithms are used to automatically extract key entities (such as equipment type, fault category, maintenance measures, etc.) and their relationships from structured and unstructured data, thereby forming a preliminary triple or semantic network structure.

[0044] Subsequently, the extracted entities and relationships are hierarchically organized and their attributes are expanded by combining domain expert knowledge with ontology design to ensure the logic and accuracy of the knowledge graph, and to enhance the semantic depth of the graph through ontology mapping.

[0045] Next, the constructed knowledge graph is imported into the visualization and reasoning engine to realize dynamic monitoring of equipment health status, intelligent detection of energy consumption anomalies, and automatic recommendation of maintenance strategies.

[0046] Finally, establish a continuous update and optimization mechanism to continuously improve and expand the knowledge graph by regularly importing the latest data, providing feedback to correct models, and introducing new knowledge sources to ensure that it always reflects the latest status and operation and maintenance needs of hotel equipment.

[0047] Through the above steps, the hotel equipment energy consumption maintenance knowledge graph not only realizes the refined management of the entire life cycle of the equipment, but also provides strong intelligent support for energy consumption optimization, fault prediction and preventive maintenance, significantly improving the hotel's operational efficiency and energy utilization.

[0048] See also Figure 2 The present invention provides a hotel energy consumption monitoring and management system based on big data, including: The correlation analysis unit collects the equipment status and environmental condition data of each area of ​​the hotel. If the weather conditions are abnormal, the energy consumption data and weather scores in the sub-area are obtained and correlation analysis is performed. The influence degree is constructed from the correlation data. , if the influence Exceeding expectations and issuing analytical instructions to external parties; The equipment optimization control unit collects the energy consumption data of each device for visualization, selects inefficient equipment based on the equipment energy consumption data and operation feedback data, takes reducing energy consumption as the control target, and uses the trained fuzzy control model to automatically control the operating parameters of inefficient equipment; The adjustment feedback unit collects the user's check-in status data, matches the user with the corresponding operation mode, collects the user's active adjustment data on the equipment in real time as feedback, optimizes the selected operation mode using the pre-trained reinforcement learning algorithm, and adds the optimized operation mode to the operation mode library; The energy consumption improvement analysis unit collects feedback data after optimization and builds improvement measures based on changes in energy consumption data. , if the improvement obtained is If the preset improvement threshold is not exceeded, a detection instruction is issued to the outside; The maintenance unit uses the operating status data of the inefficient equipment as input, detects equipment faults and issues alarm information when a fault occurs. The alarm density in the hotel area is calculated based on the alarm related information. , if the alarm density obtained If it exceeds expectations, the hotel equipment energy consumption maintenance knowledge graph will output the corresponding energy consumption optimization plan and constrain the maintenance cycle.

[0049] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0050] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0051] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only some logical function divisions. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0052] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0053] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. Hotel energy consumption monitoring and management system based on big data, characterized by: include, The correlation analysis unit collects the equipment status and environmental condition data of each area of ​​the hotel. If the weather conditions are abnormal, the energy consumption data and weather scores in the sub-area are obtained and correlation analysis is performed. The influence degree is constructed from the correlation data. , if the influence Exceeding expectations and issuing analytical instructions to external parties; The equipment optimization control unit collects the energy consumption data of each device for visualization, selects inefficient equipment based on the equipment energy consumption data and operation feedback data, takes reducing energy consumption as the control target, and uses the trained fuzzy control model to automatically control the operating parameters of inefficient equipment; The adjustment feedback unit collects the user's check-in status data, matches the user with the corresponding operation mode, collects the user's active adjustment data on the equipment as feedback in real time, optimizes the selected operation mode using the pre-trained reinforcement learning algorithm, and adds the optimized operation mode to the operation mode library; The energy consumption improvement analysis unit collects feedback data after optimization and builds improvement measures based on changes in energy consumption data. , if the improvement obtained is If the preset improvement threshold is not exceeded, a detection instruction is issued to the outside; The maintenance unit uses the operating status data of the inefficient equipment as input, detects equipment faults and issues alarm information when a fault occurs. The alarm density in the hotel area is calculated based on the alarm related information. , if the alarm density obtained If it exceeds expectations, the hotel equipment energy consumption maintenance knowledge graph will output the corresponding energy consumption optimization plan and constrain the maintenance cycle.

2. The hotel energy consumption monitoring and management system based on big data according to claim 1 is characterized in that: A sensor network is installed in the hotel area to collect equipment operating status data and environmental condition data in real time. The collected weather data is used as input, and the trained weather anomaly recognition model is used to identify anomalies and determine the degree of weather anomalies. If an anomaly exists, a weather score is output to evaluate the current weather anomaly.

3. The hotel energy consumption monitoring and management system based on big data according to claim 2 is characterized in that: Construct an electronic map covering the hotel area, divide the hotel area into several sub-areas, mark the sub-areas with correlation data and construct an influence , as follows: ； Where: , , i is the sub-region number, ; To calculate the obtained i The closeness centrality of each sub-region is For the i The correlation coefficient of the sub-regions, is the corresponding mean.

4. The hotel energy consumption monitoring and management system based on big data according to claim 3 is characterized in that: After receiving the analysis instruction, the energy consumption data of each device is recorded in real time and classified by region or device type. And record the power consumption and usage time of the equipment; The collected equipment energy consumption data and operation feedback data are used as input, and the trained energy consumption identification model is used to evaluate the equipment energy consumption. If the corresponding evaluation result does not exceed the expectation, the corresponding equipment will be marked as inefficient equipment; the location information of the inefficient equipment is obtained, and the operating parameters of the inefficient equipment are collected.

5. The hotel energy consumption monitoring and management system based on big data according to claim 4 is characterized in that: Establish an intelligent platform in the check-in area, connect the guest room equipment to the intelligent platform and deploy a dynamic sensing network to obtain the user's check-in status data and collect the user's check-in preference data through the dynamic sensing network; Design several operation modes for the equipment in the occupancy area and generate an operation mode library after summarizing them; Taking the user's check-in preference data, external weather condition data and check-in status data as input, a content-based recommendation algorithm matches the user with a corresponding operating mode from the operating mode library.

6. The hotel energy consumption monitoring and management system based on big data according to claim 5 is characterized in that: Collect user check-in feedback data within a preset feedback cycle, use the feedback data as input again, and use the pre-trained Bayesian update algorithm to automatically optimize the control parameters of the current operating mode to obtain the optimized mode; Collect the energy consumption data of equipment in the optimized occupied area within the preset observation period, and build improvement measures based on the changes in energy consumption data. , as follows: ； Where: is the starting time point of the observation period, T is the end time of the observation period, For time t The baseline energy consumption at the time, For time t Optimized energy consumption at all times, is the rate of change of energy consumption over time; is the time decay parameter, , is the weight coefficient, .

7. The hotel energy consumption monitoring and management system based on big data according to claim 6 is characterized in that: Collect the operating status data of various inefficient devices in real time, use the device operating data as input, and use the trained fault detection model to perform fault detection. If the device operating status is abnormal, it will be treated as an abnormal device and an alarm command will be issued to the outside. When the number of consecutive alarm commands received in the current stage exceeds expectations, the relevant alarm information will be summarized, and the alarm density in the hotel area will be calculated based on the alarm-related information. ; If the alarm density obtained exceeds expectations, the abnormal equipment will be marked on the electronic map, and the pre-trained path planning will plan a maintenance path for the abnormal equipment based on the location information and degree of abnormality of the abnormal equipment.

8. The hotel energy consumption monitoring and management system based on big data according to claim 7 is characterized in that: Calculate the alarm density in the hotel area based on alarm related information The way is as follows: ； Where: A is the hotel area, T To observe the cycle length, is the starting time point of the observation period; For the hotel's space domain, For the location x and time t The degree of abnormality of the equipment at the location; For the location x and time t Weather condition ratings for is the adjustment coefficient, , is the time decay parameter.

9. The hotel energy consumption monitoring and management system based on big data according to claim 8 is characterized in that: After collecting the detection data and energy consumption data of abnormal equipment, feature extraction is performed to obtain the corresponding maintenance features. The abnormal energy consumption maintenance of hotel electrical equipment is used as the target word, and the knowledge graph of hotel equipment energy consumption maintenance is pre-built; According to the correspondence between maintenance characteristics and energy consumption optimization plans, the hotel equipment energy consumption maintenance knowledge map outputs the corresponding energy consumption optimization plan, executes the energy consumption optimization plan to maintain abnormal equipment and constrains the maintenance interval .

10. The hotel energy consumption monitoring and management system based on big data according to claim 9 is characterized in that: After a maintenance interval that meets the constraints After that, maintenance constraints are applied to abnormal equipment. The constraint methods are as follows: ； Where: For maintenance intervals, T is the current time point, t is the time variable, For time t The device abnormality score at the moment, For time t Weather condition rating at the moment, is the weather condition influence coefficient, ; is the time decay parameter, is the cumulative abnormality threshold. When this threshold is reached or exceeded, maintenance is triggered. Minimum maintenance intervals to ensure that maintenance operations are not too frequent. .

Images