Electric power marketing service platform based on Internet of Things technology

Abstract

The invention relates to an electricity marketing service platform based on the Internet of Things technology. The electricity marketing service platform comprises an Internet of Things data collection perception layer, a data layer, a data model support layer and an application layer, the data model supporting layer comprises a user power consumption behavior model; the Internet of Things data collection perception layer is used for receiving the original power data of the user acquired by the acquisition equipment; the data processing module is used for preprocessing the original power data to generate effective power data; the user electricity consumption behavior model is used for generating a user electricity consumption behavior analysis result according to the user electricity consumption data; the user power consumption behavior analysis result comprises a user power consumption behavior type, a potential power consumption demand and power consumption trend prediction; the application layer matches a power marketing scheme according to the user power consumption behavior analysis result; the electricity marketing scheme comprises a personalized electricity price package and energy-saving equipment recommendation. By adopting the method and the device, the corresponding electricity marketing scheme can be matched according to the user electricity consumption behavior analysis result, the user can obtain services better meeting requirements, and the satisfaction degree is improved.

Description

Technical Field

[0001] The present invention belongs to the field of power marketing technology, and in particular relates to a power marketing service platform based on Internet of Things technology. Background Art

[0002] With the development of technology in the power industry and the opening of the power market, power marketing technology has emerged. It collects users' electricity consumption data and recommends electricity price packages to users based on the electricity consumption data.

[0003] In traditional technology, data collection mainly relies on manual meter reading. After obtaining user electricity consumption data, power companies rely on the experience of power sales personnel or simple statistical models to conduct analysis and then formulate electricity price packages for power marketing.

[0004] However, in terms of data collection, the above method has a low frequency of manual meter reading, cannot obtain users' electricity consumption information in real time, and is difficult to detect power equipment failures and abnormal power consumption in a timely manner. Simple statistics cannot predict circuit loads, cannot provide a reference for power dispatch, and cannot meet the diverse needs of different users. Users are not satisfied with electricity packages, and it is difficult for power companies to improve their market competitiveness. Summary of the Invention

[0005] Based on this, it is necessary to address the above technical problems and provide an Internet of Things technology-based power marketing service platform that can use Internet of Things devices to collect power data in real time and use big data analysis and algorithm models to analyze user power consumption behavior and load forecasting, thereby providing power marketing solutions and power scheduling solutions.

[0006] In the first aspect, this application provides a power marketing service platform based on Internet of Things technology, including:

[0007] IoT data collection perception layer, data layer, data model support layer and application layer;

[0008] The data layer includes data processing modules and data storage modules; the data model support layer includes user electricity behavior models;

[0009] The IoT data collection and perception layer is used to receive the user's original power data obtained by the collection equipment;

[0010] The data processing module is used to pre-process the original power data and generate effective power data; the effective power data includes user power consumption data;

[0011] The data storage module includes a distributed database and a relational database; the distributed database is used to store original power data; the relational database is used to store effective power data;

[0012] The user electricity consumption behavior model is used to generate user electricity consumption behavior analysis results based on user electricity consumption data; the user electricity consumption behavior analysis results include user electricity consumption behavior type, potential electricity demand and electricity consumption trend forecast;

[0013] The application layer matches electricity marketing plans based on the analysis results of users' electricity usage behavior; electricity marketing plans include personalized electricity price packages and energy-saving equipment recommendations.

[0014] In one embodiment, the user electricity usage behavior model includes a cluster analysis module, a Markov chain module, and a trend prediction module;

[0015] The cluster analysis module is used to determine the user's electricity consumption behavior type based on the user's electricity consumption data; the user's electricity consumption behavior type includes high energy consumption users, medium energy consumption users and low energy consumption users;

[0016] The Markov chain module is used to determine the transition probability of user electricity consumption behavior based on the user's electricity consumption behavior type, generate a transition probability matrix, and determine the potential electricity demand based on the transition probability matrix;

[0017] The trend prediction module is used to extract time series features based on user electricity consumption data, obtain long-term dependencies of user electricity consumption behaviors, and obtain electricity consumption trend predictions based on the long-term dependencies.

[0018] In one embodiment, determining the user's electricity usage behavior transition probability based on the user's electricity usage behavior type and generating a transition probability matrix includes:

[0019] Generate the transition probability matrix using the following formula:

[0020]

[0021] Where P is the transition probability matrix; high energy consumption users, medium energy consumption users and low energy consumption users are represented by S1, S2 and S3 respectively; p ij The user's electricity behavior type is S i Transfer to S j The probability of N ij The user's electricity behavior type is S i Transfer to S j the number of times; The user's electricity behavior type is S i The total number of times the electricity consumption behavior type is transferred to other users; i is the electricity consumption behavior type of the starting user, which is 1, 2, and 3 respectively.

[0022] In one embodiment, the data model support layer further includes a load forecasting model;

[0023] The load forecasting model is used to generate load forecast results based on effective power data;

[0024] The application layer generates the corresponding power dispatch plan based on the load forecast results and power consumption trend forecast; the power dispatch plan includes real-time dispatch plan, demand dispatch plan and economic dispatch plan.

[0025] In one embodiment, generating a load forecast result based on the effective power data includes:

[0026] The load value is generated using the following formula as the load forecast result:

[0027] δ(B)(1-B) d Y t =θ(B)∈ t

[0028] δ(B)=1-δ1B-δ2B 2 -…-δ p B p

[0029] θ(B)=1+θ1B+θ2B 2 +…+θ q B q

[0030] where Y t represents the load value at time t; B is the lag operator; (1-B) d represents the d-order difference; δ(B) is the autoregressive polynomial; θ(B) is the sliding average polynomial; ∈ t is a white noise sequence; p, d and q are model parameters.

[0031] In one embodiment, the collection equipment includes a smart meter, a concentrator, and a sensor;

[0032] Smart meters are used to collect users' electricity consumption data; electricity consumption data includes electricity consumption, voltage, current, and power;

[0033] Sensors are used to monitor the real-time operating status of power equipment and obtain power equipment operating data;

[0034] The concentrator is used to perform data verification and format conversion on the power consumption data and power equipment operation data locally to generate original power data.

[0035] In one embodiment, preprocessing raw power data to generate effective power data includes:

[0036] Perform data statistics on the original power data by user, time and region to obtain dimensional vector data;

[0037] The dimensional vector data is calculated to obtain effective power data; the effective power data includes the daily average power consumption including the time and space dimensions, the monthly cumulative power consumption including the time and space dimensions, and the average power including the time and space dimensions;

[0038] Use the following formula to get the average daily electricity consumption:

[0039]

[0040] Among them E d represents the average daily electricity consumption; P i represents the power at the i-th sampling moment; Δt i represents the i-th sampling interval;

[0041] Use the following formula to get the monthly cumulative electricity consumption:

[0042]

[0043] Among them E m Indicates the monthly cumulative electricity consumption, calculated based on a 30-day month; E d Indicates daily electricity consumption;

[0044] The average power is obtained using the following formula:

[0045]

[0046] in Indicates average power; P i represents the power at the i-th sampling moment; n is the number of sampling points.

[0047] In one embodiment, the power marketing service platform based on Internet of Things technology also includes a user layer for determining the corresponding platform interaction interface according to different user types; user types include power enterprise users, power users and cooperative users.

[0048] In one embodiment, different platform interaction interfaces are determined based on different user types, including:

[0049] When the user type is a power enterprise user, the platform interactive interface is used for users to create power marketing plans, obtain power equipment operation data, and visualize effective power data;

[0050] When the user type is an electricity user, the platform interactive interface is used for users to query user electricity consumption data, participate in electricity marketing activities corresponding to the electricity marketing plan, and handle electricity business;

[0051] When the user type is a cooperative user, the platform interaction interface is used for users to report the energy-saving equipment supply list.

[0052] In one embodiment, the application layer is further configured to generate visualized user electricity consumption data in the form of charts for query by electricity users; the visualized electricity consumption data includes daily average electricity consumption and monthly cumulative electricity consumption.

[0053] The above-mentioned power marketing service platform based on Internet of Things technology obtains the user's original power data by receiving and collecting equipment to improve the comprehensiveness and accuracy of data collection. The distributed database in the data storage module stores the original power data to ensure data integrity and traceability. The relational database stores effective power data, making the data easy to query and analyze efficiently. The user's power behavior model generates user power behavior analysis results based on effective power data. By classifying user power behavior, targeted strategies are formulated, and potential power demand is explored, services can be arranged in advance to meet user's personalized needs. Power consumption trend prediction helps users to plan power consumption reasonably, and also provides a reference for power generation and dispatching of power companies, solving the problems of imbalance in power supply and demand and difficulty in accurately grasping user needs. According to the results of user power behavior analysis, the corresponding power marketing plan is matched to improve the service quality and market competitiveness of power companies, and users obtain services that better meet their needs and improve their satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0055] Figure 1 This is an application environment diagram of the power marketing service platform based on Internet of Things technology;

[0056] Figure 2 This is a structural diagram of the power marketing service platform based on Internet of Things technology. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0058] The power marketing service platform based on Internet of Things technology provided in the embodiment of the present application can be applied to Figure 1In the application environment shown, terminal 101 communicates with server 102 via a network to support efficient data transmission. Terminal 101 can be, but is not limited to, various personal computers, laptops, smartphones, tablet computers, and IoT devices. IoT devices can include smart meters, concentrators, power equipment, and sensors. Server 102 can be implemented as a standalone server or a server cluster consisting of multiple servers.

[0059] In an exemplary embodiment, Figure 2 As shown, a power marketing service platform based on IoT technology is provided, applicable to the terminals / servers in the aforementioned implementation environment. The platform comprises an IoT data collection and perception layer, a data layer, a data model support layer, and an application layer. The IoT data collection and perception layer is primarily deployed on the terminals, and IoT devices such as smart meters, concentrators, and sensors in the terminals are all part of the IoT data collection and perception layer. The data layer is primarily deployed on the server to perform more advanced data management operations. The data model support layer is primarily deployed on the server to build complex data models for in-depth data analysis and mining. The application layer is primarily deployed on the terminal to provide various user-facing interactive interfaces and functions.

[0060] The data layer includes a data processing module and a data storage module, and the data model support layer includes a user electricity consumption behavior model.

[0061] S11. The IoT data collection and perception layer is used to receive the user's original power data obtained by the collection equipment.

[0062] The IoT's data collection and perception layer is the foundation for data reception. Receiving and collection equipment monitors electricity users' electricity usage in real time. Its efficient operation relies on the stability of the collection equipment and the real-time nature of data transmission. Wired communications such as Ethernet and fiber optics, or wireless communications, can be used to transmit data with the collection equipment at high speed and stability, enabling monitoring of raw electricity data for users within a specific area. Collection equipment can include smart meters and smart sockets installed on high-power appliances. These smart sockets can monitor appliances' real-time power consumption, operating hours, and start and stop times, enabling a more detailed analysis of residents' electricity usage.

[0063] S12. The data processing module is used to pre-process the original power data to generate effective power data, which includes user power consumption data.

[0064] The received raw power data is cleaned, converted, and integrated to extract valuable information. For example, duplicate, erroneous, and invalid data is removed from the raw power data, and the raw power data is converted to a uniform format. Using big data technology, the data is analyzed and mined in real time to provide data support for the platform's various business applications. Specifically, user electricity consumption data includes data such as the user's electricity consumption, electricity consumption time, and electricity power, which can reflect the user's basic electricity consumption behavior and electricity demand, as well as data such as peak and valley periods, average electricity load, and electricity fluctuation rate, which reflect the user's electricity consumption patterns and habits.

[0065] S13. The data storage module includes a distributed database and a relational database. The distributed database is used to store original power data, and the relational database is used to store effective power data.

[0066] The data storage module is responsible for storing data in the database. Optionally, the power data can be stored in shards to ensure efficient reading and writing of the data and long-term retention. The data storage module supports fast data retrieval. For example, when a power company needs to query the power consumption data history of a certain user, the data storage module can return the result within milliseconds. The power marketing service platform based on Internet of Things technology will continuously collect massive amounts of raw power data. By using a distributed database, the storage capacity and storage capabilities can be expanded by adding nodes to adapt to the growth of data volume and meet the long-term development needs of the platform. The raw power data is stored in multiple nodes. The failure of a single node will not result in data loss. The data copies stored in other nodes can continue to provide data services, reducing data risks. Distributed databases are conducive to large-scale data statistics in power marketing. Schematically, electricity consumption data in different regions and time periods are processed in parallel to quickly obtain results and improve data processing efficiency. By using a strict business processing mechanism stipulated by the relational database to ensure the consistency and integrity of the data, the effective power data in the relational database can be directly involved in subsequent models and applications. The stored data has high reliability and accuracy, that is, it has a clear structure and pattern, which is convenient for complex related queries and data updates, and provides a basis for subsequent users to query user electricity consumption data. In addition, the data can be overwritten within the set time, and the data within a certain time period is stored according to the setting. The storage demand is small and it is convenient for subsequent links to quickly obtain data.

[0067] S14. The user electricity usage behavior model is used to generate user electricity usage behavior analysis results based on user electricity usage data. The user electricity usage behavior analysis results include user electricity usage behavior type, potential electricity demand and electricity usage trend prediction.

[0068] The user electricity behavior model captures the patterns and characteristics of user electricity usage through in-depth analysis and mining of user electricity usage data, generating user electricity behavior analysis results with power marketing value. In principle, user electricity behavior can be classified based on similarities. Users can be divided into high, medium, and low energy consumption levels based on data characteristics such as daily electricity consumption and peak hours. Users can also be divided into daytime, nighttime, or balanced users based on the distribution of electricity usage time. This allows for a clear understanding of the electricity usage characteristics of different types of users and provides a data basis for power companies to develop differentiated marketing strategies. Analyze users' historical electricity usage data to identify changing trends and patterns in electricity demand. Based on the seasonal cycle and the stable use of common high-power electrical appliances, a forecast of future user electricity usage trends can be summarized.

[0069] S15. The application layer matches the power marketing plan based on the analysis results of the user's power consumption behavior. The power marketing plan includes personalized power price packages and energy-saving equipment recommendations.

[0070] Electricity marketing plans can be formulated by power companies based on factors such as power generation costs, transmission costs and market competition. For example, peak-valley electricity price packages are higher electricity prices during peak electricity consumption and lower electricity prices during low electricity consumption; specific time period discount packages are daytime discount packages for users whose peak electricity consumption is concentrated in the daytime; electricity gradient packages are segmented charges based on the degree of electricity consumption for users with higher electricity consumption.

[0071] For example, through the analysis results of the user's electricity usage behavior, it is found that the user's electricity usage behavior type is a high-energy-consuming user, and the peak electricity consumption is concentrated at night. The user's main energy-consuming equipment is an old, high-energy-consuming refrigerator. In this case, the personalized electricity price package in the electricity marketing plan is a peak-valley electricity price package with preferential electricity prices at night. At the same time, a refrigerator with better energy-saving effects is recommended to the user as an energy-saving device, which can not only meet the user's electricity needs, but also help the user reduce electricity costs and improve energy utilization efficiency.

[0072] Furthermore, electricity price packages can be recommended based on user behavior. This means similar price packages can be recommended for users with the same type of behavior. Furthermore, based on electricity usage trends, more favorable price packages that align with future trends can be recommended, thereby improving the market competitiveness of power companies through power marketing.

[0073] The above-mentioned power marketing service platform based on Internet of Things technology uses Internet of Things technology to collect user electricity consumption data in real time, and uses the model of the data model support layer to conduct in-depth analysis of user electricity consumption data, accurately grasp the user's electricity consumption behavior type, potential electricity demand and electricity consumption trend, and match the user's electricity consumption behavior analysis results with the power marketing plan, making marketing activities more targeted, avoiding the blindness of traditional marketing, and improving the utilization rate of marketing resources, thereby improving the economic benefits of the enterprise, and providing a basis for power companies to plan power generation plans and power dispatch, and promote the rational allocation of resources.

[0074] In one embodiment, the user electricity usage behavior model includes a cluster analysis module, a Markov chain module, and a trend prediction module;

[0075] S21. The cluster analysis module is used to determine the user's electricity consumption behavior type based on the user's electricity consumption data. The user's electricity consumption behavior type includes high energy consumption users, medium energy consumption users and low energy consumption users.

[0076] The cluster analysis module uses the K-Means algorithm to divide users into high-energy-consuming users, medium-energy-consuming users, and low-energy-consuming users. For example, the monthly cumulative electricity consumption in the user's electricity consumption data is used as the main clustering feature. The monthly cumulative electricity consumption can better reflect the user's long-term electricity consumption level. It can also be combined with the daily maximum electricity consumption, peak-to-valley ratio, etc. to more comprehensively describe the user's electricity consumption behavior type. Furthermore, the K value of the K-Means algorithm is determined to be 3 according to demand, and the users are divided into high-energy-consuming users, medium-energy-consuming users, and low-energy-consuming users. Three initial cluster centers are selected based on historical data experience. These three initial cluster centers represent the initial estimated positions of the three types of users in the distribution space of user electricity consumption data. The specific user electricity consumption data uses the Euclidean distance formula to calculate the distance from each user data point to the three cluster centers. The Euclidean distance formula is: Where x represents user electricity usage data, y represents the initial cluster center, n represents the number of features, and i represents the chronological index variable. If only monthly cumulative electricity usage is considered, n = 1; if daily peak electricity usage is also considered, n = 2. The user's electricity usage behavior type is determined by assigning the user's electricity usage data to the cluster center with the closest cluster. When matching electricity marketing plans, power companies can recommend high-energy consumption packages for high-energy users, offering more discounted electricity to meet their electricity needs and generate more revenue for the power company. They can also recommend energy-saving equipment, such as energy-saving air conditioners and water heaters, and provide purchase subsidies or rental services to help users reduce energy consumption and electricity costs. For medium-energy users, cost-effective packages can be recommended, offering appropriate price discounts while ensuring electricity quality. For low-energy users, flexible low-energy consumption packages can be recommended to meet their low electricity consumption needs, while also offering energy-saving incentives such as points redemption gifts and electricity bill discounts to encourage users to maintain energy-saving habits.

[0077] S22. The Markov chain module is used to determine the user's electricity consumption behavior transition probability according to the user's electricity consumption behavior type, generate a transition probability matrix, and determine the potential electricity demand based on the transition probability matrix.

[0078] A Markov chain module is used to characterize the potential electricity demand of potential users. Based on the transition probabilities, a transition probability matrix is ​​generated for each user's electricity behavior type. Based on the user's current electricity behavior type and the transition probability matrix, the likelihood of the user transitioning to a different electricity behavior type at the next moment, i.e., their potential electricity demand, can be inferred. For example, when the probability of a user's electricity behavior type shifting from a high-energy consumer to a medium-energy consumer is highest in the transition probability matrix, their potential electricity demand is considered medium-energy. The platform can then recommend corresponding electricity marketing solutions to encourage energy conservation.

[0079] S23, the trend prediction module is used to extract time series features based on user electricity consumption data, obtain long-term dependencies of user electricity consumption behaviors, and obtain electricity consumption trend predictions based on the long-term dependencies.

[0080] The trend prediction module is equipped with a long short-term memory network to capture the time-varying electricity consumption characteristics of user electricity consumption data in units of days, months, and years, such as the time of daily peak and trough electricity consumption, and the monthly electricity consumption patterns, to determine the long-term dependencies of user electricity consumption behaviors. The trend prediction module predicts the user electricity consumption trend for a period of time in the future. The electricity consumption trend prediction results include information such as the changing trend of electricity consumption, the time of peak and trough electricity consumption, etc.

[0081] In one embodiment, determining the user's electricity usage behavior transition probability based on the user's electricity usage behavior type and generating a transition probability matrix includes:

[0082] Generate the transition probability matrix using the following formula:

[0083]

[0084] Where P is the transition probability matrix; high energy consumption users, medium energy consumption users and low energy consumption users are represented by S1, S2 and S3 respectively; p ij The user's electricity behavior type is S i Transfer to S j The probability of N ij The user's electricity behavior type is S i Transfer to S j the number of times; The user's electricity behavior type is S i The total number of times the electricity consumption behavior type is transferred to other users; i is the electricity consumption behavior type of the starting user, which is 1, 2, and 3 respectively.

[0085] For example, the number of times a user transfers from a high energy consumption user S1 to a medium energy consumption user S2 is N 12 is 20 times, and the total number of times that high energy consumption user S1 transfers to other states is It is 100 times, then

[0086] In one embodiment, the data model support layer further includes a load forecasting model;

[0087] Using the load forecasting model, we can make short-term, medium-term and long-term forecasts of power load. Based on the load forecast results, we can make preparations for power dispatch and supply in advance and optimize the allocation of power resources. At the same time, we can also guide users to reasonably adjust their power usage time based on the load forecast results through price incentives such as electricity price packages, reduce peak loads, and improve the operating efficiency of the power system.

[0088] Load forecasting models can be trained based on a variety of data. For example, historical load data—that is, electricity load data from a long period of time, including load values ​​at different time scales—can be used. Historical load data can reflect how load changes over time, such as peaks and valleys in daily load curves or seasonal load fluctuations. Load forecasting model parameters, including the autoregressive order, differencing order, and sliding average order, are determined using methods such as maximum likelihood estimation of historical load data.

[0089] S31. The load forecasting model is used to generate load forecasting results based on effective power data.

[0090] The load forecasting model uses the autoregressive integrated moving average method in time series analysis to predict load. The autoregressive term can capture the impact of past load values ​​on current load. Load forecast results include short-term load forecast results and medium- and long-term load forecast results. Short-term load forecast results can predict load changes for the next few hours to days. Short-term load forecast results are generally used by power companies to rationally arrange the start and stop and output of generator sets. Medium- and long-term load forecast results can predict load changes for the next few months to years. Power companies generally rely on medium- and long-term load forecast results to plan and build power sources.

[0091] S32. The application layer generates a corresponding power dispatch plan based on the load forecast results and the power consumption trend forecast. The power dispatch plan includes a real-time dispatch plan, a demand dispatch plan, and an economic dispatch plan.

[0092] When short-term load forecasts and power consumption trend projections indicate increased power consumption during peak load periods, real-time dispatch plans can recommend increasing generator output during these periods to ensure adequate power supply. Power companies receiving these real-time dispatch plans can then preemptively activate backup generators to meet peak load demand. Conversely, reducing generator output during low-load periods can prevent energy waste. Real-time dispatch plans help dispatchers adjust power flow distribution in a timely manner to ensure the safe and stable operation of the power grid. When actual load deviates from forecasted load, dispatchers can balance power supply and demand by adjusting generator output or altering power flow on transmission lines.

[0093] If the medium- and long-term load forecast results and electricity consumption trends predict that future electricity demand will show an upward trend, the demand dispatch plan can recommend an increase in energy input based on the predicted continued growth in electricity demand in a certain region in the next few years. After receiving the demand dispatch plan, power companies can decide whether to build new power plants, expand existing power plants or introduce new energy power generation equipment.

[0094] An economic dispatch plan can be generated based on the load forecast results and electricity consumption trend forecast. The economic dispatch plan meets electricity demand at the lowest cost and can minimize power generation costs by optimizing the combination and output distribution of generator sets. For example, when the load demand of the load forecast results and electricity consumption trend forecast is met, clean energy units such as hydropower and wind power with lower power generation costs are generated first, and then coal-fired power is supplemented according to demand.

[0095] The use of load forecasting models is conducive to real-time understanding of power load changes, timely adjustment of power generation equipment output and allocation of power resources, and avoidance of voltage fluctuations and power outages caused by supply and demand imbalances, thereby improving the service quality and market competitiveness of power companies.

[0096] In one embodiment, generating a load forecast result based on the effective power data includes:

[0097] The load value is generated using the following formula as the load forecast result:

[0098] δ(B)(1-B) d Y t =θ(B)∈ t

[0099] δ(B)=1-δ1B-δ2B 2 -…-δ p B p

[0100] θ(B)=1+θ1B+θ2B 2 +…+θ q B q

[0101] where Y t represents the load value at time t; B is the lag operator; (1-B) d represents the d-order difference; δ(B) is the autoregressive polynomial; θ(B) is the sliding average polynomial; ∈ t is a white noise sequence; p, d and q are model parameters.

[0102] In one embodiment, the collection equipment includes a smart meter, a concentrator, and a sensor.

[0103] S41. Smart meters are used to collect users' electricity consumption data such as electricity consumption, voltage, current, and power.

[0104] Smart meters record users' electricity consumption, voltage, current, power and other electricity data in real time. The electricity consumption data is collected at certain time intervals to reflect the user's electricity consumption intensity in different time periods.

[0105] S42. The sensor is used to monitor the real-time operating status of the power equipment and obtain the operating data of the power equipment.

[0106] Temperature, humidity, vibration and other sensors are deployed on power equipment to monitor the operating status of power equipment in real time and provide data support for equipment fault warning and maintenance. Only when the power equipment is operating normally, the original power data obtained is real and usable. At the same time, the data monitored by the sensors is passed to the IoT data collection and perception layer. Power companies can quickly locate fault points based on abnormal power equipment operating data, thereby improving service quality.

[0107] S43. The concentrator is used to perform data verification and format conversion on the power consumption data and power equipment operation data locally to generate original power data.

[0108] The concentrator is mainly responsible for data aggregation and transmission to the IoT data collection and perception layer. The concentrator is connected to IoT devices such as smart meters and sensors to collect data. After collecting the data, the concentrator will perform preliminary aggregation and organization, such as data verification and format conversion, to improve the quality and efficiency of data transmission.

[0109] In one embodiment, preprocessing raw power data to generate effective power data includes:

[0110] S51. Perform data statistics on the original power data according to user, time and region dimensions to obtain dimensional vector data.

[0111] Indicatively, in the user dimension, the original power data generated by the same user at different times and on different power-consuming equipment are summarized and recorded. In the time dimension, the power data are statistically analyzed at different time scales such as day, month, season, and year, so as to analyze the change pattern of the daily point curve based on the power consumption in each statistical period. In the regional dimension, the power consumption and average power consumption of each region are statistically divided according to different geographical regions or different power lines, which is conducive to comparing the power consumption in different regions and analyzing regional power consumption characteristics, such as the differences in power consumption in commercial areas, residential areas, and industrial areas.

[0112] S52. Calculate the dimensional vector data to obtain effective power data, where the effective power data includes daily average power consumption including time and space dimensions, monthly cumulative power consumption including time and space dimensions, and average power including time and space dimensions.

[0113] Effective power data not only includes power-related data, but also includes the time and regional labels corresponding to the data, providing data support for subsequent predictions of models such as user power consumption behavior.

[0114] Use the following formula to get the average daily electricity consumption:

[0115]

[0116] Among them E d represents the average daily electricity consumption; P i represents the power at the i-th sampling moment; Δt i represents the i-th sampling interval;

[0117] Use the following formula to get the monthly cumulative electricity consumption:

[0118]

[0119] Among them E m Indicates the monthly cumulative electricity consumption, calculated based on a 30-day month; E d Indicates daily electricity consumption;

[0120] The average power is obtained using the following formula:

[0121]

[0122] in Indicates average power; P i represents the power at the i-th sampling moment; n is the number of sampling points.

[0123] In one embodiment, the power marketing service platform based on Internet of Things technology also includes a user layer, which is used to determine the corresponding platform interaction interface according to different user types, including power enterprise users, power users and cooperative users.

[0124] Power enterprise users can use the corresponding platform interactive interface to implement platform system maintenance, data maintenance, and marketing activity planning functions, understand the operating status of the power system and user electricity consumption, and provide data support for power enterprises.

[0125] Electricity users can query electricity consumption data, handle business, participate in marketing activities, and enjoy convenient and efficient electricity services through the corresponding platform interactive interface.

[0126] Partners can share data and conduct business cooperation with power companies through the corresponding platform interactive interface.

[0127] Different user types entering different platform interaction interfaces can effectively improve user convenience and the targeted nature of business behavior.

[0128] In one embodiment, different platform interaction interfaces are determined based on different user types, including:

[0129] S61. When the user type is an electric power enterprise user, the platform interactive interface is used for the user to create an electric power marketing plan, obtain electric power equipment operation data, and visualize effective electric power data.

[0130] Power companies can establish power marketing plans through an IoT-based power marketing service platform. These plans can customize electricity pricing packages based on user behavior and location, including pricing and timeframes. Power companies can also access historical operating data for power equipment such as substations, analyze equipment trends, and identify anomalies. By visualizing power data, power companies can display electricity usage by user in different regions, providing a more intuitive understanding of power distribution.

[0131] S62. When the user type is an electricity user, the platform interaction interface is used for the user to query the user's electricity consumption data, participate in the electricity marketing activities corresponding to the electricity marketing plan, and handle electricity consumption business.

[0132] When querying user electricity usage data, the IoT-based power marketing service platform provides daily, monthly, and annual timescales, displaying electricity consumption, usage distribution, and electricity costs for specific periods. Based on the recommended power marketing plan, power users can participate and receive personalized electricity price packages during the promotion period, as well as purchase energy-saving equipment.

[0133] S63. When the user type is a cooperative user, the platform interactive interface is used for the user to report the energy-saving equipment supply list.

[0134] The cooperative user may be an energy-saving equipment supplier cooperating with the power company. The energy-saving equipment supplier can provide an energy-saving equipment supply list to include information on each energy-saving equipment, including equipment name, model, energy-saving effect and price, etc. The high-energy-consuming equipment and electricity consumption links in the user's home are analyzed based on the user's electricity consumption behavior analysis results obtained from the user's electricity consumption behavior model, and energy-saving equipment that can replace high-energy-consuming equipment is recommended to the user. For example, when the user's electricity consumption behavior analysis results show that the user's electricity consumption type is a high-energy-consuming user, and the electricity consumption surges during high temperatures and non-working hours, an energy-saving air-conditioner with a higher energy efficiency rating is recommended to the user. Marketing recommendations are made based on a comparison of the energy-saving effect of the energy-saving air-conditioner, the user's electricity consumption and the price of the energy-saving air-conditioner.

[0135] In one embodiment, the application layer is further configured to generate visualized user electricity consumption data in the form of charts for query by electricity users, wherein the visualized electricity consumption data includes daily average electricity consumption and monthly cumulative electricity consumption.

[0136] For example, the user's electricity consumption data may be presented in the form of a line graph, a bar graph, a pie chart, or other charts.

[0137] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0138] The above-described embodiments merely represent several implementation methods of the embodiments of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that a person skilled in the art may make various modifications and improvements without departing from the concept of the embodiments of the present application, and these modifications and improvements fall within the scope of protection of the embodiments of the present application.

Claims

1. A power marketing service platform based on Internet of Things technology, characterized by: include: IoT data collection perception layer, data layer, data model support layer and application layer; The data layer includes a data processing module and a data storage module; The data model support layer includes a user electricity consumption behavior model; The IoT data collection and perception layer is used to receive the original power data of the user acquired by the collection device; The data processing module is used to pre-process the raw power data to generate effective power data; The effective power data includes user power consumption data; The data storage module includes a distributed database and a relational database; the distributed database is used to store the original power data; The relational database is used to store the effective power data; The user electricity usage behavior model is used to generate a user electricity usage behavior analysis result based on the user electricity usage data; The user electricity consumption behavior analysis results include user electricity consumption behavior type, potential electricity demand and electricity consumption trend forecast; The application layer matches an electricity marketing plan according to the analysis results of the user's electricity consumption behavior; the electricity marketing plan includes personalized electricity price packages and energy-saving equipment recommendations.

2. The platform according to claim 1, characterized in that: The user electricity consumption behavior model includes a cluster analysis module, a Markov chain module, and a trend prediction module; The cluster analysis module is used to determine the user's electricity consumption behavior type based on the user's electricity consumption data; the user's electricity consumption behavior type includes high energy consumption users, medium energy consumption users and low energy consumption users; The Markov chain module is used to determine the user's electricity consumption behavior transition probability according to the user's electricity consumption behavior type, generate a transition probability matrix, and determine the potential electricity demand according to the transition probability matrix; The trend prediction module is used to extract time series features based on user electricity consumption data, obtain long-term dependencies of user electricity consumption behaviors, and obtain electricity consumption trend predictions based on the long-term dependencies.

3. The platform according to claim 2, characterized in that The determining of the user's electricity usage behavior transition probability according to the user's electricity usage behavior type and generating a transition probability matrix includes: Generate the transition probability matrix using the following formula: Where P is the transition probability matrix; high energy consumption users, medium energy consumption users and low energy consumption users are represented by S1, S2 and S3 respectively; p ij The user's electricity behavior type is S i Transfer to S j The probability of N ij The user's electricity behavior type is S i Transfer to S j the number of times; The user's electricity behavior type is S i The total number of times the electricity consumption behavior type is transferred to other users; i is the electricity consumption behavior type of the starting user, which is 1, 2, and 3 respectively.

4. The platform according to claim 1, characterized in that: The data model support layer also includes a load forecasting model; The load forecasting model is used to generate a load forecasting result according to the effective power data; The application layer generates a corresponding power dispatch plan according to the load forecast result and the power consumption trend forecast; The power dispatching plan includes a real-time dispatching plan, a demand dispatching plan and an economic dispatching plan.

5. The platform according to claim 4, characterized in that Generating a load forecast result according to the effective power data includes: The load value is generated using the following formula as the load forecast result: d(B)(1-B) d Y t =θ(B)∈ t δ(B)=1-δ1B-δ2B 2 -…-d p B p θ(B)=1+θ1B+θ2B 2 +…+θ q B q where Y t represents the load value at time t; B is the lag operator; (1-B) d represents the d-order difference; δ(B) is the autoregressive polynomial; θ(B) is the sliding average polynomial; ∈ t is a white noise sequence; p, d and q are model parameters.

6. The platform according to claim 1, characterized in that: The collection equipment includes smart meters, concentrators and sensors; The smart meter is used to collect the user's electricity consumption data; the electricity consumption data includes electricity consumption, voltage, current, and power; The sensor is used to monitor the real-time operating status of the power equipment and obtain the operating data of the power equipment; The concentrator is used to perform data verification and format conversion on the power consumption data and power equipment operation data locally to generate original power data.

7. The platform according to claim 1, characterized in that The pre-processing of the raw power data to generate effective power data includes: Performing data statistics on the original power data according to user, time and region dimensions to obtain dimensional vector data; The dimensional vector data is operated to obtain effective power data; the effective power data includes daily average power consumption including time and space dimensions, monthly cumulative power consumption including time and space dimensions, and average power including time and space dimensions; Use the following formula to get the average daily electricity consumption: Among them E d represents the average daily electricity consumption; P i represents the power at the i-th sampling moment; Δt i represents the i-th sampling interval; Use the following formula to get the monthly cumulative electricity consumption: Among them E m Indicates the monthly cumulative electricity consumption, calculated based on a 30-day month; E d Indicates daily electricity consumption; The average power is obtained using the following formula: in Indicates average power; P i represents the power at the i-th sampling moment; n is the number of sampling points.

8. The platform according to any one of claims 1 to 7, characterized in that: The power marketing service platform based on Internet of Things technology also includes a user layer, which is used to determine the corresponding platform interaction interface according to different user types; the user types include power enterprise users, power users and cooperative users.

9. The platform according to claim 8, characterized in that Determine different platform interaction interfaces based on different user types, including: When the user type is a power enterprise user, the platform interaction interface is used for the user to create the power marketing plan, obtain the power equipment operation data and visualize the effective power data; When the user type is an electricity user, the platform interaction interface is used for the user to query the user's electricity consumption data, participate in the electricity marketing activities corresponding to the electricity marketing plan, and handle electricity business; When the user type is a cooperative user, the platform interaction interface is used for the user to report the energy-saving equipment supply list.

10. The platform according to claim 9, characterized in that: The application layer is further configured to generate visualized user electricity consumption data in the form of a chart for query by the power user; the visualized electricity consumption data includes the daily average electricity consumption and the monthly cumulative electricity consumption.

Images