Customer electricity consumption behavior intelligent analysis system for electricity marketing

Abstract

The invention discloses a customer electricity consumption behavior intelligent analysis system for electricity marketing, and relates to the technical field of intelligent electricity consumption analysis in the electric power industry, and the system comprises the steps of 1, user electricity consumption behavior analysis, 2, user electricity consumption scheduling analysis, and 3, user electricity consumption bidding scheduling. The power consumption behavior of the user is analyzed through four dimensions of time sequence coincidence characteristic analysis, equipment-level energy efficiency portrait, external response elasticity evaluation and power grid interaction risk early warning, and the power consumption behavior of the user can be comprehensively analyzed; by analyzing the electricity consumption behavior of the user, the electric energy which can be called or the electric energy which needs to be called can be analyzed, and then statistical classification is carried out, so that the electricity consumption behavior of the user can be better guided, and the feasibility of electricity marketing is improved; and user electricity bidding scheduling: according to the user electricity scheduling analysis in the step 2, bidding can be carried out on the users needing to be called in, so that intelligent electricity utilization is realized, a market mechanism is increased, and electricity calling is more standard and marketized.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent electricity consumption analysis in the power industry, and in particular to an intelligent analysis system for customer electricity consumption behavior oriented to power marketing. Background Art

[0002] Driven by the "dual carbon" goals and the accelerated development of new power systems, the global smart electricity market is experiencing exponential growth. According to 2024 data from the National Energy Administration, smart meter coverage in my country exceeded 95%, generating over 2.8ZB of electricity usage data annually, providing the underlying support for electricity behavior analysis.

[0003] Currently, the analysis of industrial users' electricity consumption behavior is severely fragmented, forming "data islands". The existing plant energy management system only focuses on optimizing electricity consumption within a single plant and lacks a cross-plant coordination mechanism.

[0004] In summary, an intelligent analysis system for customer electricity consumption behavior for power marketing is designed. Summary of the Invention

[0005] In order to overcome the above-mentioned deficiencies, the present invention provides an intelligent analysis system for customer electricity usage behavior for power marketing.

[0006] The present invention achieves the above-mentioned purpose through the following technical solutions: An intelligent analysis system for customer electricity consumption behavior for power marketing, comprising the following steps: Step 1: Analyze user electricity usage behavior. This involves analyzing user electricity usage behavior across four dimensions: timing compliance analysis, device-level energy efficiency profiling, external response resilience assessment, and grid interaction risk warning. Step 2: User electricity dispatch analysis: By analyzing user electricity usage behavior, we can analyze the electricity that can be called or the electricity that needs to be called in, and then perform statistical classification; Step 3: User electricity consumption bidding and scheduling. According to the user electricity consumption scheduling analysis in step 2, users who need to be transferred in can bid to achieve intelligent electricity consumption.

[0007] Preferably, the specific analysis steps of the timing compliance characteristic analysis are as follows: S11. Peak-valley characteristic analysis: Calculate the user's monthly load rate and daily peak-valley difference. If the monthly average load / maximum load is ≤ 0.7, it is inefficient; if the daily peak-valley difference is > 40%, optimization is required. S12. Production correlation analysis: Using a dynamic time warping algorithm to match production schedules, the user's load peaks and valleys are calculated. For example, a three-shift factory may have three load peaks. S13. Typical day cluster analysis uses the K-shape algorithm to identify special day patterns during holidays or peak production seasons, which is used to conduct real-time analysis of users' electricity consumption during holidays or peak production seasons; S14. Non-linear load characteristic analysis: statistics on the power consumption time and power of the user's non-linear load, and statistics on the changes in the user's power consumption after the non-linear load is working. For example, the harmonic distortion rate of the arc furnace is >30% after operation; the power factor of the variable frequency motor fluctuates between 0.2 and 0.8 after operation; the 5th / 7th / 11th harmonics of the rectifier equipment exceed the standard after operation. If there are too many non-linear loads, resulting in excessive harmonics in the power grid, it will affect the user's dispatchable rating.

[0008] Preferably, the device-level energy efficiency portrait is mainly an analysis of the electricity consumption of user equipment, which is mainly divided into three dimensions: production equipment, lighting system equipment and fire-fighting equipment. When the user is not working, the electricity consumption of production equipment and lighting system equipment can be called, but the fire-fighting equipment cannot be called.

[0009] Preferably, the specific analysis steps of the external response elasticity assessment are as follows: S15. Photovoltaic adaptability analysis: Statistical analysis of the user's photovoltaic power station power generation and usage. Since the photovoltaic power station's power generation shows a curve change, and the user's factory electricity consumption cannot be perfectly matched, the photovoltaic power station's power generation can be called upon. S16. Energy storage configuration analysis: Analyze the user's energy storage configuration. When there is sufficient energy storage power, it can be called up. When the energy storage configuration can still continue to store energy, it can call up the electricity of other users for energy storage.

[0010] Preferably, the grid interaction risk warning includes a distribution system health analysis, which is used to analyze the reliability of the user's distribution system. The main contents include risk indicator analysis of voltage sag, transformer overload and three-phase imbalance. Voltage sag is detected by a power quality monitor, which needs to be <0.9pu for 5 cycles; transformer overload is detected by optical fiber temperature measurement + load prediction; three-phase imbalance is detected by zero-sequence current analysis of smart meters.

[0011] Preferably, the user electricity scheduling analysis in step 2 includes the following specific steps: S21. Classify the types of electricity used by users into those that can be dispatched and those that need to be transferred in; S22. Analysis of the power consumption that can be dispatched or needs to be transferred in, recording the power consumption that can be dispatched by the dispatchable users and the power consumption that needs to be transferred in by the users on an hourly basis; S23. Rating of users that can be dispatched or need to be dispatched: Based on the device-level energy efficiency profile, external response resilience assessment, and grid interaction risk warning in step 1, rating of users that can be dispatched or need to be dispatched is performed; S24. Analysis of electricity prices for users who can be dispatched or need to be dispatched, and prices for users who can be dispatched or need to be dispatched to sell or buy electricity.

[0012] As a preference, the user bidding scheduling system also adds a payment system and a credit system. The payment system can realize real-time payment, and the credit system can query the operation status of the enterprise to carry out long-term cooperation.

[0013] The beneficial effects of the present invention are as follows: in the customer electricity consumption behavior intelligent analysis system for electricity marketing: 1. Through timing compliance analysis, device-level energy efficiency profiling, external response elasticity assessment, and grid interaction risk warning, the system analyzes users' electricity usage behavior from four dimensions, enabling comprehensive analysis of users' electricity usage behavior. 2. By analyzing user electricity usage behavior, we can identify the amount of electricity that can be used or that needs to be transferred in, and then statistically classify it, which can better guide user electricity usage behavior and improve the feasibility of power marketing; 3. User electricity bidding and dispatching: Based on the user electricity dispatching analysis in step 2, users who need to be transferred in can bid to realize intelligent electricity use, increase the market mechanism, make electricity dispatching more standardized and market-oriented, and more in line with user electricity needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The present invention will now be described by way of example with reference to the accompanying drawings, in which: Figure 1 It is a step diagram of the present invention. DETAILED DESCRIPTION

[0015] The present invention will now be described in further detail with reference to the accompanying drawings, which are simplified schematic diagrams that illustrate the basic structure of the present invention in a schematic manner.

[0016] like Figure 1 As shown, a customer electricity usage behavior intelligent analysis system for power marketing includes the following steps: Step 1: Analyze user electricity usage behavior. This involves analyzing user electricity usage behavior across four dimensions: timing compliance analysis, device-level energy efficiency profiling, external response resilience assessment, and grid interaction risk warning. Step 2: User electricity dispatch analysis: By analyzing user electricity usage behavior, we can analyze the electricity that can be called or the electricity that needs to be called in, and then perform statistical classification; Step 3: User electricity consumption bidding and scheduling. According to the user electricity consumption scheduling analysis in step 2, users who need to be transferred in can bid to achieve intelligent electricity consumption.

[0017] Specifically, the specific analysis steps of the timing compliance characteristic analysis are as follows: S11. Peak-valley characteristic analysis: Calculate the user's monthly load rate and daily peak-valley difference. If the monthly average load / maximum load is ≤ 0.7, it is inefficient; if the daily peak-valley difference is > 40%, optimization is required. S12. Production correlation analysis: Using a dynamic time warping algorithm to match production schedules, the user's load peaks and valleys are calculated. For example, a three-shift factory may have three load peaks. S13. Typical day cluster analysis uses the K-shape algorithm to identify special day patterns during holidays or peak production seasons, which is used to conduct real-time analysis of users' electricity consumption during holidays or peak production seasons; S14. Non-linear load characteristic analysis: statistics on the power consumption time and power of the user's non-linear load, and statistics on the changes in the user's power consumption after the non-linear load is working. For example, the harmonic distortion rate of the arc furnace is >30% after operation; the power factor of the variable frequency motor fluctuates between 0.2 and 0.8 after operation; the 5th / 7th / 11th harmonics of the rectifier equipment exceed the standard after operation. If there are too many non-linear loads, resulting in excessive harmonics in the power grid, it will affect the user's dispatchable rating.

[0018] Specifically, the device-level energy efficiency portrait mainly analyzes the electricity consumption of user equipment, which is mainly divided into three dimensions: production equipment, lighting system equipment, and fire-fighting equipment. When the user is not working, the electricity consumption of production equipment and lighting system equipment can be called, but the fire-fighting equipment cannot be called.

[0019] Specifically, the specific analysis steps of the external response resilience assessment are as follows: S15. Photovoltaic adaptability analysis: Statistical analysis of the user's photovoltaic power station power generation and usage. Since the photovoltaic power station's power generation shows a curve change, and the user's factory electricity consumption cannot be perfectly matched, the photovoltaic power station's power generation can be called upon. S16. Energy storage configuration analysis: Analyze the user's energy storage configuration. When there is sufficient energy storage power, it can be called up. When the energy storage configuration can still continue to store energy, it can call up the electricity of other users for energy storage.

[0020] Specifically, the grid interaction risk warning includes a distribution system health analysis, which is used to analyze the reliability of the user's distribution system. The main contents include risk indicator analysis of voltage sag, transformer overload and three-phase imbalance. Voltage sag is detected by a power quality monitor, which needs to be <0.9pu for 5 cycles; transformer overload is detected by optical fiber temperature measurement + load prediction; three-phase imbalance is detected by zero-sequence current analysis of smart meters.

[0021] Specifically, the user electricity scheduling analysis in step 2 includes the following specific steps: S21. Classify the types of electricity used by users into those that can be dispatched and those that need to be transferred in; S22. Analysis of the power consumption that can be dispatched or needs to be transferred in, recording the power consumption that can be dispatched by the dispatchable users and the power consumption that needs to be transferred in by the users on an hourly basis; S23. Rating of users that can be dispatched or need to be dispatched: Based on the device-level energy efficiency profile, external response resilience assessment, and grid interaction risk warning in step 1, rating of users that can be dispatched or need to be dispatched is performed; S24. Analysis of electricity prices for users who can be dispatched or need to be dispatched, and prices for users who can be dispatched or need to be dispatched to sell or buy electricity.

[0022] Specifically, the user bidding system also adds a payment system and a credit system. The payment system can realize real-time payment, and the credit system can query the operation status of the enterprise to carry out long-term cooperation. Implementation Case 1: Implementation case for manufacturing plant users (industrial application scenario).

[0023] Background description: User type: Medium-sized metal processing plant, production equipment includes electric arc furnace (non-linear load), variable frequency motor and lighting system.

[0024] Basic electricity usage: Annual electricity consumption is approximately 5 million kWh, with a low monthly load factor (average load / maximum load = 0.6) and a daily peak-to-valley variation of up to 50%. Production operates on a three-shift basis, and electricity consumption increases significantly during holidays.

[0025] System implementation goal: By analyzing user electricity consumption behavior, the factory's dispatchable electricity is called upon during off-peak hours to optimize the grid load.

[0026] Step 1: Analyze user electricity usage behavior Timing compliance characteristic analysis: S11. Peak-valley characteristic analysis: Analyze the monthly load rate (0.6<0.7, marked as inefficient) and the daily peak-valley difference (50%>40%, needs to be optimized).

[0027] S12. Production correlation analysis: Use the dynamic time warping algorithm to match the production schedule and identify load peaks (three peaks corresponding to the three shift times: 8:00-10:00, 14:00-16:00, 22:00-24:00) and valleys (2:00-6:00 in the morning).

[0028] S13. Typical day cluster analysis: Use the K-shape algorithm to identify holiday patterns (e.g., electricity consumption drops by 30% during the Spring Festival shutdown period).

[0029] S14. Non-linear load characteristics analysis: Statistics on the harmonic distortion rate (>30%) when the arc furnace is working, which causes power factor fluctuations (0.2~0.8) and affects the scheduling rating.

[0030] Equipment-level energy efficiency portrait: mainly analyzes production equipment (electric arc furnaces, variable frequency motors, which can be used during non-working hours), lighting systems (can be used), and fire-fighting equipment (cannot be used).

[0031] External response resilience assessment: does not involve photovoltaics and energy storage.

[0032] Grid interaction risk warning: Analyze voltage sag (<0.9pu for 5 cycles, detected by power quality monitor) and three-phase imbalance (zero-sequence current exceeds the standard).

[0033] Step 2: User electricity dispatch analysis S21. Classification of electricity consumption types: Identified as “dispatchable user electricity consumption” (dispatchable electricity can be provided during off-peak hours).

[0034] S22. Electricity consumption analysis: record the dispatchable electricity by hour (50kWh / h can be dispatched before 6:00 during off-peak hours).

[0035] S23. User rating: Based on equipment-level profiling (production equipment has high adjustability), external elasticity assessment (no elasticity) and risk warning (moderate risk of three-phase imbalance), the rating is B (moderately dispatchable).

[0036] S24 price analysis: The user sets the electricity selling price to 0.4 yuan / kWh (lower than the peak price).

[0037] Step 3: User bidding scheduling Bidding mechanism: Users who need to be dispatched (such as nearby commercial buildings) bid to purchase dispatchable electricity.

[0038] Payment system: Realize real-time transaction payment (such as completing transfers during off-peak hours).

[0039] Credit reporting system: check the credit of factories and enterprises (good) and support long-term cooperation.

[0040] Expected results Business advantages: The factory generates revenue by using off-peak electricity (saving 12% of electricity bill annually), and the peak-to-valley difference of the power grid is reduced to 30%.

[0041] Technical implementation: Based on the dynamic time warping algorithm to predict the load, the system automatically triggers the scheduling process.

[0042] Case value: Verify the system's effectiveness in solving the "data island" problem of industrial users and improve collaborative scheduling efficiency.

[0043] Implementation Case 2: Implementation Case for Commercial Building Users (Commercial Application Scenario) Background Description User type: Modern office building, main equipment includes lighting system, air conditioning and fire-fighting equipment.

[0044] Basic electricity usage: Daily peak-to-valley power consumption varies by 40%. The load is highest during working hours (9:00 AM - 6:00 PM), while lighting and air conditioning can be adjusted during non-working hours. Voltage sags occur frequently (twice per month).

[0045] System implementation objectives: Analyze the required power during peak hours and ensure the safety of the fire protection system.

[0046] Step 1: Analyze user electricity usage behavior Timing compliance characteristic analysis: S11. Peak-valley characteristic analysis: daily peak-valley difference is 40% (critical point needs to be optimized).

[0047] S13. Typical day cluster analysis: Identify weekend low-load patterns (40% drop in electricity consumption).

[0048] Production relevance and non-linear loads (non-industrial users) are not involved.

[0049] Equipment-level energy efficiency portrait: lighting system equipment (available during non-working hours), fire-fighting equipment (not available); analysis of the available ratio (lighting system accounts for 30% of electricity consumption).

[0050] External response resilience assessment: does not involve photovoltaics.

[0051] Grid interaction risk warning: voltage sag (<0.9pu for 5 cycles) and transformer overload risk (predicted by fiber optic temperature measurement).

[0052] Step 2: User electricity dispatch analysis S21. Classification of electricity consumption types: Identify as “users who need to transfer in electricity” (electricity needs to be transferred in during peak hours).

[0053] S22. Electricity consumption analysis: record the amount of electricity required by the hour (100kWh / h is required during peak hours 9:00-11:00).

[0054] S23. User rating: Based on the equipment-level portrait (the risk of increased non-adjustability of fire-fighting equipment) and risk warning (risk of temporary voltage increase), the rating is C (priority for adjustment).

[0055] S24. Price analysis: The user sets the electricity purchase price to 0.8 yuan / kWh (high price during peak hours).

[0056] Step 3: User bidding scheduling Bidding mechanism: Dispatched users (such as the factory in Implementation Case 1) bid to provide electricity.

[0057] Payment system: Instant processing of transactions (payment via APP).

[0058] Credit reporting system: Check the credit of the building management company (no arrears record) and facilitate long-term contract agreements.

[0059] Expected results Business advantages: The building reduced peak electricity costs by 18% (cited in an International Energy Agency report) and prevented power outages through risk early warning.

[0060] Technical implementation: The K-shape algorithm optimizes holiday scheduling, and the credit reporting system reduces transaction risks.

[0061] Case value: reflects the application of the system in the commercial field to solve the "complexity of new energy access" and strengthen the integration of payment and credit reporting.

[0062] Implementation Case 3: Implementation Case of Photovoltaic Integration Users (New Energy Application Scenario) Background Description User type: Food processing plant, equipped with a photovoltaic power station and energy storage system, and production equipment including production lines and fire protection systems.

[0063] Basic electricity usage: PV generation is mismatched with user demand (high generation during the day but low demand), requiring energy to be deployed when sufficient storage is available. There are few nonlinear loads, and the power distribution is healthy.

[0064] System implementation goal: Analyze dispatchable photovoltaic power and realize competitive bidding.

[0065] Step 1: Analyze user electricity usage behavior Timing compliance characteristic analysis: S11. Peak-valley characteristic analysis: monthly load rate 0.75 (high efficiency), daily peak-valley difference 30% (no optimization required).

[0066] Omit production relevance (non-three-shift operation).

[0067] Equipment-level energy efficiency profile: production equipment (callable), fire-fighting equipment (not callable).

[0068] External Response Resilience Assessment: S15, Photovoltaic adaptability analysis: Power generation (peak 500kW) exceeds plant usage (300kW), and the 200kW difference can be adjusted.

[0069] S16. Energy storage configuration analysis: The energy storage device has a surplus of 50kWh / h of electricity available between 3:00 and 5:00 p.m.; or when energy storage is available, external electricity is used for charging.

[0070] Grid interaction risk warning: three-phase balance is good (zero-sequence current is lower than the risk threshold).

[0071] Step 2: User electricity dispatch analysis S21. Classification of electricity consumption: Identified as “dispatchable user electricity” (surplus electricity from photovoltaic and energy storage).

[0072] S22. Electricity consumption analysis: record the dispatchable electricity by hour (200kWh / h can be dispatched during the photovoltaic adaptation period).

[0073] S23. User rating: Based on external elasticity assessment (photovoltaic adaptability and high energy storage elasticity) and device-level profiling (low risk), the rating is A (excellent dispatchable).

[0074] S24. Price analysis: The user sets the electricity selling price at 0.3 yuan / kWh (low price to attract electricity).

[0075] Step 3: User bidding scheduling Bidding mechanism: Users who need to be dispatched (such as the building in Implementation Case 2) bid to purchase dispatchable photovoltaic power.

[0076] Payment system: real-time settlement (blockchain technology enhances security).

[0077] Credit reporting system: Check the factory credit (good credit), used for multiple transactions.

[0078] Expected results Business advantage: The factory can increase its annual revenue by 15% by selling excess photovoltaic power, thereby improving the utilization rate of new energy (refer to the complexity of distributed photovoltaics mentioned in the background of the text).

[0079] Technical implementation: The S15 photovoltaic adaptability analysis algorithm automatically optimizes calls, and the payment system reduces transaction delays.

[0080] Case value: Demonstrate the application of the system under the "dual carbon goals" and verify the system's innovation through cross-user collaborative scheduling.

[0081] Three implementation cases demonstrate how the system solves practical electricity consumption problems (such as industrial valley call, commercial peak call, and new energy collaboration), and combines payment and credit reporting systems to achieve intelligent bidding.

[0082] The above description is for inspiration. Based on the above description, relevant personnel can make various changes and modifications without departing from the technical concept of this invention. The technical scope of this invention is not limited to the content of the specification, but must be determined according to the scope of the claims.

Claims

1. An intelligent analysis system for customer electricity usage behavior for power marketing, characterized by: The following steps are involved: Step 1: Analyze user electricity usage behavior. This involves analyzing user electricity usage behavior across four dimensions: timing compliance analysis, device-level energy efficiency profiling, external response resilience assessment, and grid interaction risk warning. Step 2: User electricity dispatch analysis: By analyzing user electricity usage behavior, we can analyze the electricity that can be called or the electricity that needs to be called in, and then perform statistical classification; Step 3: User electricity consumption bidding and scheduling. According to the user electricity consumption scheduling analysis in step 2, users who need to be transferred in can bid to achieve intelligent electricity consumption.

2. The intelligent analysis system for customer electricity usage behavior for power marketing according to claim 1 is characterized by: The specific analysis steps of the timing compliance characteristic analysis are as follows: S11. Peak-valley characteristic analysis: statistics on the user's monthly load rate and daily peak-valley difference; S12. Production correlation analysis: using a dynamic time warping algorithm to match production schedules and calculate user load peaks and valleys; S13, typical day cluster analysis, using the K-shape algorithm to identify special day patterns during holidays or peak production seasons; S14. Non-linear load characteristic analysis: statistics on the power consumption time and power of the user's non-linear load, and statistics on the changes in the user's power consumption after the non-linear load is working.

3. The intelligent analysis system for customer electricity usage behavior for electricity marketing according to claim 1 is characterized by: The device-level energy efficiency portrait mainly analyzes the electricity consumption of user equipment, which is mainly divided into three dimensions: production equipment, lighting system equipment, and fire-fighting equipment.

4. The customer electricity usage behavior intelligent analysis system for power marketing according to claim 1 is characterized by: The specific analysis steps of the external response resilience assessment are as follows: S15, Photovoltaic adaptability analysis, statistical analysis of the user's photovoltaic power station power generation and usage; S16. Energy storage configuration analysis: Analyze the user's energy storage configuration. When there is sufficient energy storage power, it can be called up. When the energy storage configuration can still continue to store energy, it can call up the electricity of other users for energy storage.

5. The intelligent analysis system for customer electricity usage behavior for electricity marketing according to claim 1 is characterized by: The grid interaction risk warning includes a distribution system health analysis, which is used to analyze the reliability of the user's distribution system. The main contents include risk indicator analysis of voltage sag, transformer overload and three-phase imbalance.

6. The customer electricity usage behavior intelligent analysis system for electricity marketing according to claim 1 is characterized by: The user electricity scheduling analysis in step 2 includes the following specific steps: S21. Classify the types of electricity used by users into those that can be dispatched and those that need to be transferred in; S22. Analysis of the power consumption that can be dispatched or needs to be transferred in, recording the power consumption that can be dispatched by the dispatchable users and the power consumption that needs to be transferred in by the users on an hourly basis; S23. Rating of users that can be dispatched or need to be dispatched: Based on the device-level energy efficiency profile, external response resilience assessment, and grid interaction risk warning in step 1, rating of users that can be dispatched or need to be dispatched is performed; S24. Analysis of electricity prices for users who can be dispatched or need to be dispatched, and prices for users who can be dispatched or need to be dispatched to sell or buy electricity.

7. The customer electricity usage behavior intelligent analysis system for electricity marketing according to claim 1 is characterized by: The user bidding system also adds a payment system and a credit system. The payment system can realize real-time payment, and the credit system can query the operation status of the enterprise to carry out long-term cooperation.

Images