User electric power actuarial system based on artificial intelligence

Abstract

The invention discloses a user electric power actuarial system based on artificial intelligence, and relates to the technical field of electric power intelligent services, and the system comprises the following components: a data collection and storage module which is connected with a power grid and user electric power equipment through an electric power sensor and a communication interface, collects data such as user electric charge, electric quantity and the like and power grid electricity price information, and transmits the data to a power supply module; storing the data in a database; data are deeply analyzed and modeled by applying a machine learning algorithm, and targeted energy-saving suggestions are generated for users by predicting future power consumption, load change and electric charge expenditure of the users and analyzing association between different power consumption behaviors and electric charges, so that not only is the actual power consumption condition of the users considered, but also the energy-saving effect of the users is improved. And information such as the real-time electricity price and the peak-valley time period of the power grid is combined, so that individuation and precision of energy-saving suggestions are realized, and a user is helped to reasonably control the electricity utilization cost and optimize the electricity utilization behavior.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent power services, and specifically provides a user power actuarial system based on artificial intelligence. Background Art

[0002] With the rapid development of society and the progress of technology, power services have become one of the indispensable infrastructures in modern society. As important components of power services, power demand forecasting and cost actuarial have a significant impact on the operation and management of power enterprises and the electricity consumption experience of users. In recent years, the rise of artificial intelligence technology has provided a new opportunity for the intelligentization and refinement of power services. By integrating user power consumption behavior data and emotional data and using artificial intelligence algorithms for in-depth analysis, it is possible to accurately predict users' power demand and cost expenditure, thereby providing users with personalized and efficient power services.

[0003] Traditional power data analysis mainly relies on manual experience and basic statistical methods. These methods are unable to cope when dealing with modern power data with complexity and variability. Especially when faced with abnormal data caused by factors such as equipment failures, human violations, and data acquisition deviations, traditional methods are difficult to effectively detect and correct, resulting in inaccurate analysis results. In addition, traditional methods lack the ability to deeply explore the value of data and are difficult to provide accurate and personalized energy-saving strategies. Therefore, users often have difficulty obtaining practical and effective energy-saving suggestions in actual applications, and the effect of optimizing electricity costs is not obvious.

[0004] In view of the above problems, it is necessary to optimize the existing user power actuarial system based on artificial intelligence. By deeply analyzing and modeling the data, it is possible to predict users' future electricity consumption, load changes, and electricity bill expenditures, and generate targeted energy-saving suggestions according to the actual electricity consumption situation. Therefore, it is of great significance to develop an artificial intelligence-based user power actuarial system that can comprehensively achieve the above characteristics. Summary of the Invention

[0005] The purpose of the present invention is to make up for the deficiencies of the existing technology and provide a user power actuarial system based on artificial intelligence. It can effectively detect and correct abnormal data by comprehensively collecting, processing, and analyzing power grid and user power-related data, combined with advanced time series analysis and artificial intelligence algorithms, improving the accuracy and reliability of data analysis. At the same time, the system can generate suitable energy-saving suggestions for users based on the analysis results, helping users optimize their electricity consumption behavior and reduce electricity costs. It not only improves the intelligent level of data analysis but also significantly enhances the pertinence and practicality of energy-saving suggestions, bringing new breakthroughs to the fields of power data analysis and energy-saving technology.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: A user power precise calculation system based on artificial intelligence, which system comprises the following components: Data acquisition and storage module: Collects data such as user electricity charges and electricity consumption, as well as grid electricity price information, through the marketing data center, and stores them in the database; Data preprocessing and anomaly detection module: Preprocesses the data in the database, uses the multi-dimensional dynamic threshold method to monitor the power consumption data in real time, identifies and analyzes anomaly points, and corrects or marks the abnormal data; Data analysis and modeling module: Analyzes the data after preprocessing and anomaly correction using a comprehensive prediction model, establishes models for predicting the user's future electricity consumption, load changes, and electricity charge expenditures, and analyzes the correlation between different electricity consumption behaviors and electricity charges; Energy-saving suggestion generation module: Based on the results of data analysis and modeling, combined with the user's actual electricity consumption situation, comprehensively considering factors such as electricity consumption changes, electricity prices, and implementation costs, uses the energy-saving benefit evaluation formula to evaluate the energy-saving benefits and generate targeted energy-saving suggestions; Interaction module: Deployed in the intranet of the power company to provide an interaction interface for the staff of the power supply company, display the energy-saving suggestion information corresponding to the user, and generate a corresponding user power precise calculation report for the staff to download, and adjust the priority of the energy-saving suggestions according to the staff feedback and usage frequency factors to optimize the interaction experience and system effect.

[0007] Further, the data preprocessing and anomaly detection module uses the multi-dimensional dynamic threshold method to monitor the power consumption data in real time, identify and analyze anomaly points, and its algorithm formula is: , where is the dynamic upper limit threshold of the th data dimension, is the dynamic lower limit threshold of the th data dimension, is the mean value of the th data dimension, is the standard deviation of the th data dimension, is the threshold coefficient of the th data dimension, adjusted according to the importance and fluctuation characteristics of the data in this dimension, is the time influence coefficient of the th data dimension, reflecting the influence of the data change trend over time on the threshold, is the time variable, representing the order or timestamp of the collected data, is the correlation influence coefficient of the th data dimension, is the comprehensive influence index of the relevant data dimension, obtained from the user power-related data stored in the database.

[0008] Furthermore, the data preprocessing and anomaly detection module corrects or marks the abnormal data. Specifically, it traverses all data points in each data dimension, compares the value of the data point with the corresponding threshold. For a data point whose value is greater than or less than , the data point is marked as an abnormal point. For abnormal data whose accurate value can be determined, it is corrected according to the cause of the anomaly. For abnormal data caused by equipment failure, the abnormal data is corrected according to the normal operating parameters of the equipment. For abnormal data caused by data acquisition deviation, the abnormal data is corrected by re-acquiring data or calibrating the acquired data. For abnormal data whose accurate value cannot be determined, it is marked as abnormal data, a marking field is added to the data, and the marking value of the abnormal data is set to a specific value.

[0009] Furthermore, the data analysis and modeling module uses machine learning techniques to deeply analyze and model the data after preprocessing and anomaly correction. By learning and analyzing historical data, a prediction model is established to predict the user's future electricity consumption, load changes, and electricity bill expenditures. At the same time, it analyzes the correlation between different electricity consumption behaviors and electricity bills, finds the key factors affecting electricity bills, continuously optimizes and adjusts the established model, evaluates the performance of the model using test data, and adjusts the parameters of the model according to the evaluation results.

[0010] Furthermore, the data analysis and modeling module establishes a prediction model through learning and analysis of historical data. The model formula is: , where is the predicted electricity consumption, is the intercept term, which is calculated by performing a regression analysis on historical data, is the number of linear influencing factors, is the th regression coefficient of the linear influencing factor, is the th linear influencing factor, that is, the linear influencing factor data stored in the database, is the number of non-linear influencing factors, is the th coefficient of the non-linear influencing factor, is the th functional transformation of the non-linear influencing factor, is the th non-linear influencing factor, is the error term, reflecting the random factors that the model cannot explain.

[0011] Further, the energy-saving recommendation generation module evaluates the energy-saving benefit using the energy-saving benefit evaluation formula, and the formula is: , where is the energy-saving benefit, that is, the amount of electricity bill saved after taking energy-saving measures, is the predicted power consumption after taking energy-saving measures, is the power consumption without taking energy-saving measures, is the average electricity price, is the evaluation time period, is the implementation cost coefficient of the energy-saving measures.

[0012] Further, the energy-saving recommendation generation module generates targeted energy-saving recommendations. For the optimization of electricity usage time, according to the peak and valley periods of the electricity price, it is recommended that users use high-power electrical equipment during the valley period and avoid using electricity during the peak period; for the adjustment of the basic electricity charge collection method, when the basic electricity charge generated by the user is higher than that of other collection methods, a suggestion is put forward to the user to select the preferred basic electricity charge collection method according to their own production situation; for the power factor adjustment, it is recommended that users take measures to improve the power factor and reduce the reactive power loss.

[0013] Further, the interaction module adjusts the priority of the energy-saving recommendations according to the staff feedback and usage frequency factors, and its priority evaluation formula is: , where is the adjusted priority of the th energy-saving recommendation, is the priority of the th energy-saving recommendation before adjustment, is the staff feedback influence coefficient, which is used to control the influence degree of staff feedback on the recommendation priority, is the staff feedback score of the th energy-saving recommendation, and the value range is [-1, 1]. A positive number indicates that the staff is satisfied with the recommendation, and a negative number indicates dissatisfaction, is the usage frequency influence coefficient, which is used to control the influence degree of the recommendation usage frequency on the priority, is the usage frequency of the th energy-saving recommendation, which is calculated by the ratio of the number of times the user uses this recommendation to the total number of times the recommendations are used.

[0014] Compared with the prior art, this user power actuarial system based on artificial intelligence has the following beneficial effects: 1. The present invention deeply analyzes and models data by applying machine learning algorithms. By predicting the user's future electricity consumption, load changes, and electricity bill expenditures, and analyzing the correlation between different electricity consumption behaviors and electricity bills, personalized energy-saving suggestions are generated for users. It not only considers the user's actual electricity consumption situation but also combines information such as the real-time electricity price and peak-valley periods of the power grid, thus realizing the personalization and precision of energy-saving suggestions, helping users reasonably control electricity costs, optimize electricity consumption behaviors, and promoting the reasonable allocation and efficient utilization of power resources, which is of great significance for promoting energy conservation, emission reduction, and sustainable development.

[0015] 2. The present invention significantly improves the intelligent level of power data processing by introducing advanced artificial intelligence technologies. By real-time monitoring power consumption data, it effectively identifies and corrects abnormal data caused by factors such as equipment failures, human violations, or data collection deviations, thus ensuring the high accuracy and reliability of the data. This not only provides a solid foundation for subsequent data analysis and modeling but also greatly improves the efficiency and precision of power data management.

[0016] Other advantages, objectives, and features of the present invention will be described to some extent in the subsequent specification, and to some extent, will be obvious to those skilled in the art based on the study of the following text, or can be learned from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or in the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings without creative efforts based on these drawings.

[0018] Figure 1 It is a schematic structural diagram of a user power actuarial system based on artificial intelligence; Figure 2 It is a flowchart of a user power actuarial system based on artificial intelligence. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] To further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention objectives, the following, in combination with the drawings and preferred embodiments, details the specific embodiments, structures, features, and their effects of the present invention as follows.

[0020] Embodiment 1 A large manufacturing electricity-consuming enterprise, mainly engaged in the production and manufacturing of auto parts, has a huge production workshop and numerous production equipment, such as stamping machines, CNC machine tools, welding equipment, etc. Its complex electricity consumption system makes the electricity cost account for a relatively large proportion in the enterprise's cost. In order to effectively control costs and improve energy utilization efficiency, through the power supply company, this user power precise calculation system based on artificial intelligence is applied to provide energy-saving suggestions for the enterprise's production.

[0021] The system collects data such as the enterprise's electricity cost, electricity consumption, load, power factor, etc. through the marketing data middle platform, and at the same time obtains information such as the grid electricity price, peak-valley periods, etc. These data are stored orderly in the system's powerful database, providing a data basis for subsequent analysis and processing.

[0022] Preprocess the collected data, use the multi-dimensional dynamic threshold method to monitor the power consumption data in real time, identify and analyze abnormal points, and conduct 24-hour real-time monitoring of the enterprise's power consumption data. Its algorithm formula is: , where is the dynamic upper limit threshold of the th data dimension, is the dynamic lower limit threshold of the th data dimension, is the mean value of the th data dimension, is the standard deviation of the th data dimension, is the threshold coefficient of the th data dimension, adjusted according to the importance and fluctuation characteristics of the data in this dimension, is the time influence coefficient of the th data dimension, reflecting the influence of the data's change trend over time on the threshold, is the time variable, representing the order or timestamp of the collected data, is the correlation influence coefficient of the th data dimension, is the comprehensive influence index of relevant data dimensions, obtained from the user power-related data stored in the database. During the monitoring process, it is found that the electricity consumption of a certain auto parts production line shows an abnormal increase during the period from 10 am to 11 am on Tuesday. The system quickly activates the abnormal cause analysis mechanism. By deeply analyzing the historical operation data of the production line equipment, including the changes in parameters such as the equipment's temperature, vibration, current, etc., and comparing the electricity consumption behavior patterns of the same type of production lines, it is determined that a key motor of this production line has failed, resulting in a decrease in the motor's operating efficiency and a significant increase in energy consumption. The system clearly marks this abnormal data and immediately notifies the staff of the power supply company in the form of text messages and in-system notifications.

[0023] 8. Use machine learning algorithms to deeply analyze and model the data after preprocessing and anomaly correction. Its model formula is: , where is the predicted electricity consumption, is the intercept term, which is calculated by performing a regression analysis on historical data, is the number of linear influencing factors, is the th regression coefficient of the linear influencing factor, is the th linear influencing factor, that is, the linear influencing factor data stored in the database, is the number of non-linear influencing factors, is the th coefficient of the non-linear influencing factor, is the th functional transformation of the non-linear influencing factor, is the th non-linear influencing factor, is the error term, which reflects the random factors that the model cannot explain. By learning and analyzing the enterprise's electricity consumption data in the past year, the system can accurately predict the enterprise's future electricity consumption, load changes, and electricity bill expenditures. After detailed analysis, it is found that the enterprise has a large electricity consumption during peak hours, and some old equipment has low operating efficiency due to backward technology, which not only increases energy consumption but also leads to high electricity bill costs. For example, some stamping equipment purchased earlier has significantly higher energy consumption than new equipment and frequent failures.

[0024] Based on the results of data analysis and modeling, combined with the enterprise's actual production situation and development plan, the system generates a series of highly targeted and practical energy-saving suggestions for the enterprise, including adjusting the operating time of some non-critical equipment and arranging it to operate during valley hours to make full use of the advantages of low valley electricity prices; suggesting to the enterprise to choose the preferred basic electricity charge collection method according to its own production situation, etc. In addition, for power factor adjustment, it is recommended that the enterprise take measures to improve the power factor and reduce reactive power loss.

[0025] Power supply company staff can view energy-saving suggestions and detailed electricity cost analysis reports through the interactive interface, deeply understand the enterprise's electricity consumption situation and energy-saving potential. According to the actual production arrangement and cost budget, the power supply company staff suggest that the enterprise make a reasonable adjustment to the electricity plan and formulate a detailed energy-saving transformation plan. Three months after implementing the energy-saving suggestions, the enterprise feeds back the implementation effect through the system and finds that the electricity consumption has decreased by 12% compared with before, and the electricity cost has been reduced by about 15%. Based on the enterprise's feedback information, the system further optimizes the energy-saving suggestions and data analysis model, continuously monitors the enterprise's electricity consumption, and provides more accurate and effective energy-saving strategies for the enterprise to achieve more efficient energy-saving effects and cost control.

[0026] Embodiment 2 An electricity-consuming enterprise in a commercial complex, located in the core business district of the city, includes various business forms such as large shopping malls, high-end office buildings, and diverse catering and entertainment venues. Its electricity consumption situation is complex and has obvious time-period and seasonal characteristics. In order to achieve the goals of energy conservation, emission reduction, and cost reduction in operation, the power supply company applies this artificial intelligence-based user power actuarial system to generate energy-saving suggestions for the enterprise.

[0027] The system establishes connections with various electricity-consuming devices in the commercial complex, such as large-area lighting systems, large central air-conditioning systems, multiple elevators, and catering and entertainment equipment, as well as the power grid system. According to the operation rules and electricity consumption characteristics of the commercial complex, the system collects data such as electricity costs, electricity consumption, load, and power factor, and at the same time obtains information such as electricity prices and peak-valley periods of the power grid, and securely stores these data in the database for in-depth analysis later.

[0028] Strict preprocessing is carried out on the stored data. Data cleaning technology is used to clean the data, removing noise and invalid information. The multi-dimensional dynamic threshold method is used to monitor the power consumption data, identify and analyze abnormal points. During the abnormal detection process, it is found that the lighting system on a certain floor has abnormal power consumption during non-business hours (from 2 am to 5 am), far higher than the normal level. The system immediately starts a detailed abnormal cause analysis process. By analyzing the operation records of relevant lighting equipment, checking the control switch status of the equipment, and comparing the electricity consumption behavior patterns of other floors, it is judged that there may be electricity theft behavior or equipment leakage problems. The system marks this abnormal data as a key point and promptly notifies the enterprise's security and power management personnel by phone and email for a comprehensive investigation. After the joint inspection by security personnel and power technicians, it is finally determined that an aging and damaged lighting line on this floor has caused the leakage phenomenon.

[0029] Using machine learning algorithms to deeply analyze and model the processed data, and combining the historical electricity consumption data and real-time operation data of the commercial complex, the system can accurately predict the future electricity consumption, load changes, and electricity bill expenditures of the commercial complex. The analysis results show that the air conditioning system in the shopping mall accounts for a relatively large proportion of electricity consumption in summer, and due to the aging of air conditioning equipment, the refrigeration efficiency has decreased and the energy consumption is relatively high. At the same time, the elevators in the office building have a high operation frequency during the peak commuting hours, and the empty running rate is also relatively high, resulting in unnecessary energy waste.

[0030] Based on the results of data analysis and modeling, and combining the operation characteristics and customer needs of the commercial complex, the system generates a series of comprehensive and targeted energy-saving suggestions for the enterprise. For example, adjust the operation time of some non-critical equipment and arrange it to operate during the valley period to make full use of the advantages of low valley electricity prices; put forward suggestions to the enterprise to choose the preferred basic electricity charge collection method according to its own production situation, etc. In addition, for power factor adjustment, it is recommended that the enterprise take measures to improve the power factor and reduce reactive power loss.

[0031] The staff of the power supply company learned about the energy-saving suggestions and electricity consumption data in detail through the interaction interface, and adjusted the operation strategy according to the actual operation situation and budget arrangement of the commercial complex. After implementing the energy-saving measures for half a year, the enterprise reported the energy-saving effect to the system. The overall electricity consumption decreased by 10% compared with before, and the electricity cost was reduced by about 13%. At the same time, the indoor environmental comfort of the shopping mall and the office building was not affected, but was improved to a certain extent due to the application of the intelligent control system. The system optimized the energy-saving suggestions and data analysis model according to the feedback information of the enterprise, continuously helped the enterprise tap the energy-saving potential, further reduced the electricity cost, improved the energy utilization efficiency, and at the same time enhanced the overall operation level and competitiveness of the commercial complex.

[0032] The above is only a preferred embodiment of the present invention, and it is not intended to limit the present invention in any form. Although the present invention has been disclosed above with the preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art can make some changes or modifications to the above-disclosed technical content to be equivalent changes within the scope of the technical solution of the present invention. However, as long as it does not depart from the content of the technical solution of the present invention, any brief modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. A user power actuarial system based on artificial intelligence, characterized in that: The system consists of the following components: Data collection and storage module: collects user electricity charges, electricity consumption data and grid electricity price information through the marketing data center and stores them in the database; Data preprocessing and anomaly detection module: preprocess the data in the database, use the multi-dimensional dynamic threshold method to monitor the power consumption data in real time, identify and analyze abnormal points, and correct or mark abnormal data; Data analysis and modeling module: Use a comprehensive prediction model to analyze pre-processed and anomaly-corrected data, establish a model to predict users' future electricity consumption, load changes, and electricity bills, and analyze the relationship between different electricity consumption behaviors and electricity bills; Energy-saving suggestion generation module: Based on the results of data analysis and modeling, combined with the actual electricity consumption of users, taking into account the changes in electricity consumption, electricity prices and implementation costs, the energy-saving benefit evaluation formula is used to evaluate the energy-saving benefits and generate targeted energy-saving suggestions; Interactive module: Deployed in the power company's intranet to provide an interactive interface for power supply company staff, display energy-saving recommendation information for corresponding users, and generate corresponding user power actuarial reports for staff to download. The priority of energy-saving recommendations is adjusted according to staff feedback and usage frequency factors to optimize the interactive experience and system effect.

2. According to the artificial intelligence-based user power actuarial system of claim 1, it is characterized in that: The data preprocessing and anomaly detection module uses a multi-dimensional dynamic threshold method to monitor power consumption data in real time, identify and analyze anomalies, and its algorithm formula is: ,in, It is The dynamic upper threshold of the data dimension, It is The dynamic lower threshold of the data dimension, It is The mean of the data dimensions, It is The standard deviation of the data dimension, It is The threshold coefficient of the data dimension is The importance and volatility characteristics of each data dimension are adjusted. It is The time impact coefficient of each data dimension reflects the impact of the data change trend over time on the threshold. is a time variable, indicating the order or timestamp of collected data, It is The correlation influence coefficient of each data dimension, It is a comprehensive impact indicator of relevant data dimensions, derived from user power-related data stored in the database.

3. According to the artificial intelligence-based user power actuarial system of claim 2, it is characterized in that: The data preprocessing and anomaly detection module corrects or marks the abnormal data. Specifically, it traverses all data points in each data dimension, compares the value of the data point with the corresponding threshold, and or less than , then mark the data point as an abnormal point. For abnormal data whose exact value can be clearly determined, make corrections based on the cause of the abnormality. For abnormal data caused by equipment failure, make corrections based on the normal operating parameters of the equipment. For abnormal data caused by data collection deviation, make corrections by recollecting the data or correcting the collected data. For abnormal data whose exact value cannot be determined, mark it as abnormal data, add a tag field to the data, and set the tag value of the abnormal data to a specific value. .

4. The user power actuarial system based on artificial intelligence according to claim 1 is characterized in that: The data analysis and modeling module uses machine learning technology to perform in-depth analysis and modeling on the pre-processed and anomaly-corrected data. By learning and analyzing historical data, a prediction model is established to predict the user's future electricity consumption, load changes, and electricity bill expenditures. At the same time, the relationship between different electricity consumption behaviors and electricity bills is analyzed to find out the key factors affecting electricity bills, and the established model is continuously optimized and adjusted. The performance of the model is evaluated by using test data, and the parameters of the model are adjusted according to the evaluation results.

5. The user power actuarial system based on artificial intelligence according to claim 4 is characterized in that: The data analysis and modeling module establishes a prediction model by learning and analyzing historical data. The model formula is: ,in, is the predicted electricity consumption, is the intercept term, which is calculated by regression analysis of historical data. is the number of linear influencing factors, It is The regression coefficient of the linear influencing factor is It is linear influencing factors, that is, the linear influencing factor data stored in the database, is the number of nonlinear influencing factors, It is The coefficients of the nonlinear influencing factors are It is The function transformation of nonlinear influencing factors, It is Nonlinear influencing factors, is the error term, reflecting random factors that cannot be explained by the model.

6. The user power actuarial system based on artificial intelligence according to claim 1 is characterized in that: The energy-saving suggestion generation module uses an energy-saving benefit evaluation formula to evaluate the energy-saving benefit, and the formula is: ,in, is the energy-saving benefit, that is, the amount of electricity saved after taking energy-saving measures, is the predicted electricity consumption after energy-saving measures are taken, is the electricity consumption when no energy-saving measures are taken. is the average electricity price, is the evaluation period, It is the implementation cost coefficient of energy-saving measures.

7. The user power actuarial system based on artificial intelligence according to claim 1 is characterized in that: The energy-saving suggestion generation module generates targeted energy-saving suggestions. For optimization of electricity consumption time, according to the peak and valley periods of electricity prices, users are advised to use high-power electrical equipment during valley periods to avoid electricity consumption during peak periods. For adjustment of the basic electricity fee calculation method, when the basic electricity fee generated by the user is higher than that of other calculation methods, the user is advised to choose the preferred basic electricity fee calculation method based on his or her own production conditions, that is, a targeted strategy based on data analysis and modeling results combined with the user's actual electricity consumption conditions. For power factor adjustment, users are advised to take measures to improve the power factor and reduce reactive power loss.

8. The user power actuarial system based on artificial intelligence according to claim 1 is characterized in that: The interactive module adjusts the priority of energy-saving suggestions based on staff feedback and usage frequency factors. The priority evaluation formula is: ,in, It is The energy saving suggestions have been prioritized. It is The priority of the energy saving suggestions before adjustment, is the staff feedback influence coefficient, which is used to control the influence of staff feedback on the priority of recommendations. It is The staff feedback score of the energy-saving suggestions is in the range of [-1, 1]. A positive number indicates that the staff is satisfied with the suggestion, and a negative number indicates that the staff is dissatisfied. It is the frequency of use influence coefficient, which is used to control the influence of the recommended frequency of use on the priority. It is The frequency of use of an energy-saving suggestion is calculated by counting the ratio of the number of times users use this suggestion to the total number of times the suggestions are used.

Images