Planned electric quantity management and control method and system based on multi-index cooperative tracking

Abstract

The invention discloses a planned electric quantity management and control method and system based on multi-index cooperative tracking, and the method belongs to the technical field of digital management and control of an electric power system, and achieves the collection and standardization of cross-system data through constructing a multi-index system comprising a plan type index, an execution type index, a prediction type index and a deviation type index; a cooperative tracking mechanism is established based on the multi-index system, hierarchical early warning is triggered by comparing execution type, plan type and prediction type indexes in real time, a supervision order is automatically generated and tracked to a deviation closed loop, and full-process management and control are achieved; and finally, a plan decomposition rule and an early warning threshold are optimized through multi-dimensional analysis, and a refined management and control suggestion is output. According to the method, the precision, efficiency and decision support capability of planned electric quantity management and control can be remarkably improved, and the method is suitable for multi-level (group, company and station) electric quantity management and control scenes of a large-scale electric power group.

Description

Technical Field

[0001] This invention relates to the field of digital management technology for power systems, and in particular to a method and system for planned power generation control based on multi-indicator collaborative tracking. It is especially suitable for the multi-level (group, company, substation) full-process control of planned power generation within large power groups. Background Technology

[0002] Under the market-oriented reforms (such as electricity spot trading) and the need for refined management in the power industry, large power groups face the following key technical challenges in controlling planned electricity volume: Data fragmentation and lack of standardization: Data related to planned electricity generation (such as power plant reports, real-time power generation, and hydrological / wind speed data) are scattered across production systems, marketing systems, and independent power plant systems. The data formats are inconsistent (e.g., power generation units include kWh and MWh), and the timestamps are inconsistent. Cross-system data integration relies on manual processes, which is inefficient, error-prone, and cannot support real-time management and control.

[0003] The plan decomposition lacks intelligent coordination: the existing plan decomposition mostly relies on manual allocation according to fixed proportions, without taking into account the characteristics of different power generation types (hydro, thermal, wind, solar) (such as hydropower being greatly affected by water inflow and wind power being greatly affected by wind speed), resulting in a mismatch between the decomposition results and the actual output capacity; moreover, the handling of tail differences during the decomposition process (such as the deviation between the total component and the total amount) requires manual adjustment, which is time-consuming and prone to errors.

[0004] Isolated monitoring indicators and delayed early warnings: Existing monitoring focuses only on the single deviation of "actual power generation - planned power generation" without coordinating with predictive indicators (such as future monthly power generation forecasts), making it impossible to predict deviation risks in advance; moreover, the early warning threshold is fixed and is not dynamically adjusted according to the type of power generation (e.g., the early warning threshold for hydropower during the wet season and the dry season should be different), resulting in delayed early warnings or false alarms.

[0005] The supervision loop lacks linkage with indicator deviations: When indicator deviations occur, the generation of supervision orders relies on manual triggering and is not automatically linked to the deviation indicators, resulting in insufficient targeting of supervision; moreover, the supervision process lacks real-time tracking (such as whether the responsible unit handles it in a timely manner and whether the handling effect meets the standards), the closed-loop management is missing, and deviation problems cannot be resolved in a timely manner.

[0006] Limited analytical dimensions and weak decision support: Existing analyses are mostly limited to "comparison between actual and planned data", lacking year-on-year analysis (comparison with the same period in history) and forecast deviation analysis (comparison between forecast and actual data); the analysis results are presented in tabular form, which cannot intuitively locate weak links in management and control (such as a certain type of power generation or a certain power plant), making it difficult to support refined decision-making.

[0007] Therefore, there is an urgent need for a planned power consumption management solution that can integrate multi-source data, coordinate multiple indicators, and achieve intelligent decomposition, real-time monitoring, supervision, and precise analysis to solve the above-mentioned technical problems. Summary of the Invention

[0008] The purpose of this invention is to propose a planned power consumption management method based on multi-indicator collaborative tracking, which aims to solve the problems of scattered planned power consumption management data, isolated indicators, and delayed decision-making in the prior art.

[0009] This invention is implemented as follows: a planned power consumption management method based on multi-indicator collaborative tracking, the method comprising: Construct a multi-indicator collaborative system; the multi-indicator collaborative system includes, but is not limited to, planning indicators, execution indicators, forecasting indicators, and deviation indicators; Data collection and standardization processing are performed; the collected data includes the core data sources of planning, execution, and forecasting indicators in the multi-indicator collaborative system, including real-time operation data, planning and transaction-related data, site reporting data, and cross-system related data; Multi-indicator collaborative tracking is implemented, and hierarchical early warnings are triggered by real-time comparison of execution, planning and forecast indicators. Supervision orders are automatically generated for deviation events, and supervision feedback is tracked until the deviation is closed. Conduct multi-dimensional analysis and optimization; based on the analysis results, optimize the plan decomposition rules and hierarchical early warning thresholds, and output corresponding control recommendations.

[0010] Another objective of this invention is to propose a planned power consumption management system based on multi-indicator collaborative tracking, the system comprising: The indicator system construction module is used to construct a multi-indicator collaborative system; the multi-indicator collaborative system includes, but is not limited to, planning indicators, execution indicators, forecasting indicators, and deviation indicators. The data acquisition and standardization module is used for data acquisition and standardization processing; the acquired data includes the core data sources of planning, execution, and forecasting indicators in the multi-indicator collaborative system, including real-time operation data, planning and transaction-related data, site reporting data, and cross-system related data; The collaborative tracking module is used for multi-indicator collaborative tracking. It triggers tiered early warnings by comparing execution, planning and forecast indicators in real time, automatically generates supervision orders for deviation events, and tracks supervision feedback until the deviation is closed. The analysis and optimization module is used for multi-dimensional analysis and optimization; based on the analysis results, it optimizes the plan decomposition rules and hierarchical early warning thresholds, and outputs corresponding control recommendations.

[0011] Beneficial effects of the present invention This invention proposes a planned power generation control method and system based on multi-indicator collaborative tracking. Belonging to the field of digital power system management technology, this method constructs a multi-indicator system encompassing planning, execution, forecasting, and deviation indicators to achieve cross-system data collection and standardization. Based on this system, a collaborative tracking mechanism is established, triggering tiered early warnings through real-time comparison of execution, planning, and forecasting indicators. This automatically generates supervision orders and tracks deviations to a closed loop, achieving full-process control. Finally, through multi-dimensional analysis and optimization of planning decomposition rules and early warning thresholds, refined management suggestions are output. This invention significantly improves the accuracy, efficiency, and decision support capabilities of planned power generation control, and is suitable for multi-level (group, company, substation) power generation control scenarios in large power groups. Attached Figure Description

[0012] Figure 1 This is a flowchart of a planned power consumption management method based on multi-indicator collaborative tracking, according to a preferred embodiment of the present invention. Figure 2 This is a structural diagram of a planned power consumption management system based on multi-indicator collaborative tracking, according to a preferred embodiment of the present invention. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. For ease of explanation, only the parts related to the embodiments of this invention are shown. It should be understood that the specific embodiments described herein are merely for explaining this invention and are not intended to limit this invention.

[0014] This invention proposes a method and system for planned power generation control based on multi-indicator collaborative tracking. The method belongs to the field of digital power system management technology. It constructs a multi-indicator system including planning, execution, forecasting, and deviation indicators to achieve cross-system data collection and standardization. Based on this multi-indicator system, a collaborative tracking mechanism is established. Real-time comparison of execution, planning, and forecasting indicators triggers tiered early warnings, automatically generates supervision orders, and tracks deviations to a closed loop, achieving full-process control. Finally, through multi-dimensional analysis and optimization of planning decomposition rules and early warning thresholds, refined management suggestions are output. This invention can significantly improve the accuracy, efficiency, and decision support capabilities of planned power generation control, and is suitable for multi-level (group, company, station) power generation control scenarios in large power groups.

[0015] Figure 1 This is a flowchart of a planned power consumption management method based on multi-indicator collaborative tracking, according to a preferred embodiment of the present invention; the method includes the following steps: S1: Construct a multi-indicator collaborative system; the multi-indicator collaborative system includes, but is not limited to, planning indicators, execution indicators, forecasting indicators, and deviation indicators; Planned indicators: These should include at least the group's annual planned electricity volume (the total annual amount allocated by the group), the company's annual or monthly planned electricity volume, and the annual or monthly planned electricity volume reported by the power stations; Implementation indicators: These should include at least the daily or monthly power generation of the power plant, cumulative power generation, and plan completion rate ((actual power generation / planned power generation) × 100%). Forecast indicators: at least include forecasted power generation based on historical averages (average power generation over the past N years, where N is selected from 3 to 10 years based on the accumulation of data from power plants) and forecasted power generation based on the characteristics of power generation type; Deviation indicators: These should include at least the deviation from the planned completion rate (the difference between the actual completion rate and the target completion rate) and the prediction deviation (the difference between the predicted electricity generation and the actual electricity generation). In this embodiment of the invention, the predicted power generation based on power generation type characteristics includes at least one of the following predicted power generation amounts: Hydropower characteristic correction prediction of power generation: The core correction characteristic is the dependence on hydrological conditions. The correction basis includes water inflow, water level and other hydrological data collected by the real-time production system, as well as hydrological-power generation correlation data for the same period in the past N years (N is selected as 3-10 years based on the accumulation of station data); The calculation logic is: average power generation in the same period of the past N years × water inflow deviation coefficient; whereby the water inflow deviation coefficient is the ratio of the actual water inflow in the current month to the average water inflow in the same period of the past N years. Thermal power characteristic correction forecast power: The core correction characteristic is load factor stability. The correction basis includes grid load forecast data, fuel supply stability data (coal / gas inventory), and load factor-power generation correlation data for the same period in the past N years (N is selected as 3-10 years based on the accumulation of data from power plants). The calculation logic is: average power generation in the same period of the past N years × load factor correction coefficient. Wherein, the load factor correction coefficient is the ratio of the grid forecast load factor for the current month to the average load factor in the same period of the past N years, and a fuel supply correction coefficient can be added according to actual demand. Wind power characteristic correction prediction: The core correction characteristic is the timeliness of wind speed. The correction is based on real-time / short-term wind speed forecasts issued by meteorological departments, measured wind speed data from wind farms, and wind speed-power generation correlation data for the same period over the past N years (N is selected as 3-10 years based on the data accumulation of the wind farms). The calculation logic is: average power generation for the same period over the past N years × wind speed correction coefficient. The wind speed correction coefficient is the ratio of the actual average wind speed of the current month to the average wind speed for the same period over the past N years. The predicted value for each time period can be further optimized according to the daily time-of-day wind speed distribution. Photovoltaic characteristic correction for predicted power generation: The core correction characteristic is the dependence of irradiance intensity on sunshine duration. The correction is based on irradiance intensity forecasts issued by meteorological departments, measured irradiance data from photovoltaic power plants, and irradiance-power generation correlation data for the same period over the past N years (N is selected as 3-10 years based on the accumulation of data from the power plants). The calculation logic is: average power generation for the same period over the past N years × (irradiance correction coefficient × sunshine duration correction coefficient); where the irradiance correction coefficient is the ratio of the actual average irradiance intensity of the current month to the average irradiance intensity for the same period over the past N years; and the sunshine duration correction coefficient is the ratio of the actual sunshine duration of the current month to the average sunshine duration for the same period over the past N years.

[0016] In this embodiment of the invention, N is selected based on the data accumulation of hydropower stations; for newly built stations, 3 years can be used, and for mature stations, 5-10 years can be used. For example, the following details the calculation process for photovoltaic characteristic correction prediction power generation in a specific application scenario: Assuming a mature photovoltaic power plant (with 7 years of data accumulation, and N chosen as 5 years) needs to calculate the photovoltaic characteristic-corrected predicted power generation in July 2025, the specific process is as follows: Data collection: The actual average irradiance in July 2025 was obtained from the meteorological department as 6.8 kWh / m², and the actual sunshine duration in July was obtained from the photovoltaic power station measurement system as 245 hours. The correlation data of irradiance and power generation in July of the past 5 years (2020-2024) were extracted from the system's historical database. The average irradiance in July of the past 5 years was 5.7 kWh / m², the average sunshine duration in July of the past 5 years was 228 hours, and the power generation in July of the past 5 years was 320 million kWh, 330 million kWh, 310 million kWh, 340 million kWh, and 300 million kWh, respectively.

[0017] Basic data calculations: Average power generation in the same period of the past 5 years = (3.2 + 3.3 + 3.1 + 3.4 + 3.0) billion kWh ÷ 5 = 3.2 billion kWh; Irradiance correction factor = Actual average irradiance intensity in July 2025 ÷ Average irradiance intensity in July of the past 5 years = 6.8 kWh / m² ÷ 5.7 kWh / m² ≈ 1.193; Sunshine duration correction factor = Actual sunshine duration in July 2025 ÷ Average sunshine duration in July of the past 5 years = 245 hours ÷ 228 hours ≈ 1.074.

[0018] Calculation of photovoltaic power generation with characteristic correction: July 2025 photovoltaic power generation with characteristic correction = average power generation of the same period in the past 5 years × irradiance correction coefficient × sunshine duration correction coefficient = 320 million kWh × 1.193 × 1.074 ≈ 3.2 × 1.281 ≈ 410 million kWh.

[0019] S2: Data acquisition and standardization processing; In this embodiment of the invention, the collected data includes: Real-time operational data: Data such as power generation, water conditions (including inflow and water level), wind speed, irradiance, sunshine duration, load factor, and equipment operating parameters can be obtained from the real-time production system via the MQTT protocol; Planning and transaction-related data: Group / company annual / monthly plan issuance data, transaction execution data, and scheduling instruction-related data can be synchronized from the marketing system and group planning system through ETL tools; Data submitted by the stations: Each station can submit its annual / monthly planned electricity consumption through a unified submission portal; Cross-system related data: electricity settlement data in the marketing system, plan decomposition benchmark data in the group planning system, and other derivative data related to planning and control.

[0020] The collected data comprehensively covers the core data sources of planning, execution, and forecasting indicators in the multi-indicator collaborative system, providing complete data support for subsequent standardized processing and full-process control. It also corresponds to the standardization requirements of unified data units and format specifications, ensuring the relevance and comprehensiveness of data collection.

[0021] In this embodiment of the invention, the collected data is standardized, and the data units are unified (e.g., power generation is unified to 10,000 kWh, water inflow is unified to m³ / s), timestamp format (e.g., unified to yyyy-MM-dd HH:mm:ss), and data encoding rules are unified (e.g., a unique code is generated according to the "group-company-station-indicator" hierarchy) to ensure that the data format is consistent across systems. In this embodiment of the invention, abnormal data during the collection process can be identified and marked using the 3σ criterion, triggering a data specialist to verify and remind the user. Missing data can be supplemented using linear interpolation. At the same time, cross-system data consistency is checked periodically (such as comparing power generation data between the production system and the marketing system), and a data quality control report is output. Finally, a complete, accurate, and unified standardized dataset is formed, laying the data foundation for subsequent multi-indicator collaborative tracking and multi-dimensional analysis.

[0022] S3: Perform multi-indicator collaborative tracking, trigger tiered early warnings by real-time comparison of execution, planning, and forecasting indicators, automatically generate supervision orders for deviation events, and track supervision feedback until the deviation loop is closed; including the following steps: S31: Based on preset electricity decomposition rules, automatically decompose the group / company's planned electricity consumption into annual monthly, monthly daily, and daily hourly electricity consumption. Furthermore, tail differences are automatically processed (e.g., tail differences are adjusted to the last non-null weight value). In this embodiment of the invention, the power decomposition rules include: configuring decomposition weights according to power generation type; hydropower can be configured with monthly weights based on the average monthly power generation ratio over the past N years (N is selected as 3-10 years based on the data accumulation of the power station); thermal power can be configured with daily time-of-use weights based on the historical load factor curve; and wind power and photovoltaic power can be configured with time-of-use weights based on the historical output curve. If a power station does not have its own decomposition rules, it can automatically inherit the general decomposition rules for its power generation type.

[0023] The historical load factor curve refers to a time-series curve generated based on the historical operating data of thermal power plants over the past N years (N is selected as 3-10 years depending on the accumulated data of the power plants) for the same period (e.g., annual / monthly / daily). The curve uses 24 time periods (e.g., 00:00-01:00 to 23:00-24:00) as the horizontal axis and the load factor of each time period as the vertical axis. The load factor is calculated as: Load Factor = Actual Power Generation Load of the Time Period ÷ Rated Power Generation Load of the Unit × 100%. This curve is generated statistically from real-time thermal power plant operating data (including unit output data, rated load parameters, and time period power generation duration) stored in the system's historical database. It intuitively reflects the load carrying capacity and operational stability distribution of thermal power plants at different times of the day.

[0024] The historical output curve refers to a time-series curve generated based on the historical operating data of wind / solar power plants over the past N years (N is selected as 3-10 years based on the accumulated data of the power plants) (e.g., annual / monthly / daily). The curve uses "time period" (e.g., every 1 hour / every 2 hours as a time period) as the horizontal axis and the actual output (unit: 10,000 kWh or kW) of the corresponding time period as the vertical axis. This curve is generated statistically from real-time wind / solar power operating data (including wind speed / irradiance correlation data, actual power generation during the time period, equipment operating status data, etc.) stored in the system's historical database. It primarily reflects the output fluctuation patterns of wind / solar power affected by natural conditions (wind speed, irradiance).

[0025] For example, the power decomposition rules are explained in detail below with reference to specific application scenarios; Suppose that the group assigns a company a planned electricity generation of 6 billion kWh for 2025 (4 billion kWh of hydropower and 2 billion kWh of thermal power). The automatic allocation is executed according to the preset electricity generation allocation rules, as follows: Configuration of general decomposition rules for power generation types: Hydropower: Monthly weighting is allocated based on the average monthly power generation ratio over the past three years (e.g., January 5%, February 6%, March 8%, April 10%, May 15%, June 14%, July 13%, August 9%, September 7%, October 6%, November 5%, December 2%); Daily time-based weighting is allocated according to the intraday water inflow pattern (e.g., 10:00-16:00 is the peak output period, with a weight of 1.2 for each time period, and 0.8 for other time periods); Thermal power: Daily time-of-use weights are allocated based on the historical load factor curves of the past three years (1.5 for peak hours 10:00-12:00 and 18:00-22:00, 1.0 for flat hours 08:00-10:00, 12:00-18:00 and 22:00-24:00, and 0.8 for off-peak hours 00:00-08:00), with a balanced monthly weight allocation (8.33% per month). Wind power: The general rule is to allocate daily time-of-day weights based on the historical output curve (14:00-18:00 peak wind speed, weight 1.5 for each time period, and weight 0.9 for other time periods), and monthly weights are allocated based on the wind resource distribution over the past three years (e.g., 10% each for March-May and September-November, 8% each for January-February and June-August, and 6% for December).

[0026] Year and Month Breakdown: Hydropower: May's allocated electricity = 4 billion kWh × 15% = 600 million kWh, June = 4 billion kWh × 14% = 560 million kWh... December = 4 billion kWh × 2% = 80 million kWh; the total for each month after allocation is 3.998 billion kWh, with a slight difference of 2 million kWh automatically adjusted to the last non-empty weighted month of December, resulting in an adjusted December electricity amount of 82 million kWh. Thermal power: Monthly decomposed electricity = 2 billion kWh × 8.33% ≈ 166.67 million kWh, with no tail difference.

[0027] Monthly and daily breakdown: Hydropower in May: 31 days in total, average daily decomposed electricity = 600 million kWh ÷ 31 ≈ 19.3548 million kWh; Thermal power in June: 30 days in total, average daily decomposed electricity = 166.67 million kWh ÷ 30 ≈ 5.5557 million kWh.

[0028] Daily and hourly breakdown: Hydropower on May 15 (normal water day): There are a total of 96 15-minute time slots. From 10:00 to 16:00 (24 time slots), the electricity consumption per time slot is approximately 19,354,800 kWh × 1.2 ÷ 96 ≈ 241,900 kWh. The electricity consumption for the remaining time slots is approximately 19,354,800 kWh × 0.8 ÷ 96 ≈ 161,300 kWh. Thermal power on June 20 (working day): 18:00-19:00 (4 time periods) Electricity per time period = 5,555,700 kWh × 1.5 ÷ 96 ≈ 87,100 kWh; 02:00-03:00 (4 time periods) Electricity per time period = 5,555,700 kWh × 0.8 ÷ 96 ≈ 46,300 kWh.

[0029] Station rules inheritance: A newly built wind farm under the company does not have its own allocation rules and automatically inherits the general allocation rules for wind power: If the group allocates a planned power generation of 200 million kWh for the wind farm in 2025, the allocated power generation in March = 200 million kWh × 10% = 20 million kWh. The average daily power generation on March 10th = 20 million kWh ÷ 31 ≈ 64,500 kWh. The power generation during the period from 14:00 to 15:00 = 64,500 kWh × 1.5 ÷ 96 ≈ 1,010 kWh.

[0030] S32: Real-time comparison of execution indicators with planning indicators and forecast indicators. When deviation indicators exceed preset thresholds (such as plan completion rate <90%), a tiered warning (such as red / yellow warning) is triggered. In this embodiment of the invention, the triggering condition for the graded early warning can be set as follows: Red alert: Plan completion rate <90%, or forecast deviation >15%; Yellow alert: 90% ≤ plan completion rate < 95%, or 10% < forecast deviation ≤ 15%; The warning information is synchronized to the work platform of the responsible unit and reaches the responsible person through message notifications (such as system pop-ups and text messages).

[0031] In one embodiment of the present invention, the method for comparing execution indicators with planning indicators and prediction indicators is as follows: Based on the unified dataset after S2 standardization, execution indicators, planning indicators and prediction indicators are synchronized in real time at a cycle consistent with the data collection frequency (≥5 minutes / time). The comparison process follows the principle of "time granularity matching + hierarchical dimension correspondence": First, according to the three-level control dimension of "group-company-station", the execution indicators (such as daily power generation and cumulative power generation of the station) of the same organizational level are associated with the planning indicators (planned power generation of the same station on the same day and cumulative planned power generation of the same period) of the corresponding period. The comparison between the two is achieved by calculating the plan completion rate (cumulative power generation / planned power generation of the corresponding period × 100%), which intuitively presents the degree of deviation between the actual execution and the plan benchmark. Then, the execution indicators (such as the monthly cumulative power generation of the power station) at the same organizational level and time granularity are correlated with the prediction indicators (the predicted power generation of the same power station in the same month based on historical averages or power generation type characteristics). The prediction deviation (predicted power generation - actual power generation) and prediction deviation rate (predicted deviation / actual power generation × 100%) are calculated to compare the two and quantify the degree of fit between the actual execution and the prediction value. The comparison results are synchronously mapped to the corresponding deviation indicators, providing direct data support for subsequent determination of whether the preset threshold is exceeded and triggering graded early warning.

[0032] For example, suppose that a company's subsidiary power plant A (thermal power) has a planned monthly electricity generation of 357 million kWh in June 2025. The system collects production data in real time and compares it with multiple indicators. The specific early warning triggering and notification process is as follows: On June 10th, the daily execution indicators (actual power generation of 1.144 million kWh) and monthly planned indicators (daily planned power generation of 1.3 million kWh) of Power Plant A were synchronized at the preset frequency. The calculated daily plan completion rate was 88% (1.144 million kWh ÷ 1.3 million kWh × 100%). This value was lower than the red alert threshold (90%), so a red alert was triggered in real time. At the same time, the June forecast indicators of Power Plant A were retrieved (monthly forecast power generation of 342 million kWh based on the historical load rate of thermal power). As of June 10th, the cumulative actual power generation was 102 million kWh, and the cumulative forecast power generation was 115 million kWh. The calculated forecast deviation rate was 11.3% ((115 - 102) ÷ 102 × 100%), which did not exceed the red alert threshold (15%). Therefore, the red alert was triggered only for the deviation in the plan completion rate. The system then automatically links the core data of the deviation event (planned completion rate 88%, daily plan 1.3 million kWh, actual 1.144 million kWh, deviation of -156,000 kWh, threshold standard 90%), and determines the responsible unit as Power Plant A and the responsible person as the marketing manager of the power plant according to the "station-company" hierarchical responsibility mechanism. A standardized early warning notification is generated simultaneously. The notification is immediately synchronized to the management and control workbench of Power Plant A (a pop-up window displays the deviation details, early warning level and handling requirements), and is also pushed to the responsible person's mobile phone via SMS.

[0033] For example, in June 2025, the monthly electricity forecast for the company's B power plant (hydropower) based on water situation correction was 620 million kWh. The actual electricity generated that month was 545.6 million kWh, and the calculated forecast deviation rate was 12% ((620 million - 545.6 million) ÷ 545.6 million × 100%). This value is in the range of 10% to 15%, triggering a yellow alert. The alert information was synchronized to the B power plant's work platform and notified to the responsible person through the APP, specifying a completion deadline of 5 working days to ensure that the responsible person received the notification in a timely manner and initiated rectification.

[0034] S33: Automatically generate a supervision order for deviation events that trigger graded early warnings, determine the responsible unit, responsible person and completion deadline, track supervision feedback (such as handling measures, completion status) until the deviation is closed; Specifically, in this embodiment of the invention, for deviation events that trigger graded early warnings, a standardized supervision order can be automatically generated based on the core data of deviation-related indicators in the multi-indicator collaborative system (including planned values, actual values, deviation values, deviation rates, and threshold standards for triggering early warnings for planned, executed, and predicted related indicators). The supervision order has a built-in three-level responsibility traceability mechanism of "group-company-station", which can automatically match the responsible unit and extract the responsible person according to the organizational level and power generation type to which the deviation event belongs. At the same time, the completion time limit can be dynamically set according to the severity of the deviation (e.g., the completion time limit for red early warning deviation events is ≤2 working days, the completion time limit for yellow early warning deviation events is ≤5 working days, and special scenarios can be manually adjusted but adjustment records must be retained). The system can reach responsible personnel via pop-up windows on the workbench, SMS messages on mobile devices, or push notifications from the app. Responsible personnel must submit targeted handling measures (such as adjustments to equipment operating parameters, communication and coordination records with the power grid dispatching department, load optimization plans, fuel supply guarantee measures, etc.) and relevant supporting materials (such as dispatching instruction receipts, equipment maintenance records, and operational data reports) through the system within the specified timeframe. The system will synchronize the completion status of each supervision order (pending / processing / pending acceptance / closed loop / overdue) in real time and display it visually on the management workbench at all levels. For supervision orders that fail to submit handling measures or fail to meet rectification requirements within the deadline, a secondary warning will be automatically triggered, and the superior responsible unit will be notified. The superior unit can then intervene to coordinate the supervision. After rectification is completed, the system can automatically collect data on execution indicators for the corresponding subsequent cycles (such as power generation and plan completion rate for 3 natural days / 1 control cycle after rectification) to verify the rectification effect. At the same time, the superior responsible unit will review and accept the compliance and effectiveness of the handling measures. Once the execution indicators return to the warning threshold range and the review is passed, the supervision order will be marked as "closed loop", ensuring that the deviation event forms a traceable and quantifiable control loop from triggering the warning to solving the problem.

[0035] S4: Conduct multi-dimensional analysis and optimization; optimize the plan decomposition rules and hierarchical early warning thresholds based on the analysis results, and output corresponding control recommendations.

[0036] In this embodiment of the invention, based on the cross-system full data after standardization processing in step S2 and the closed-loop control data formed by multi-indicator collaborative tracking in step S3, multi-dimensional analysis is performed based on the linkage relationship between planning, execution, prediction, and deviation indicators in the multi-indicator collaborative system, including: Year-on-year analysis: By annual / monthly / daily granularity, the performance indicators of the same period this year are compared with the data of the same period in the past N years (N is selected from 3 to 10 years based on the accumulation of site data) to quantify the historical operating trend; Planned comparison analysis: Based on the three-level management dimension of "group-company-station", the actual cumulative power generation at each level is compared with the corresponding planned indicators for the period to identify the level at which the plan execution deviation is attributable and the weak links in coordination. Prediction deviation analysis: Classified by hydropower, thermal power, wind power, and photovoltaic power, calculate the deviation value and deviation rate between the predicted power generation (including predicted power generation based on historical averages and predicted power generation based on power generation type characteristics) and the actual power generation, evaluate the accuracy of different prediction logics, and especially conduct quantitative verification of the correction effect for different power generation type characteristics. Based on the above analysis results, the planning decomposition rules for each type of power generation can be optimized and adapted accordingly (such as adjusting the weighting of hydropower peak and off-peak periods, the weighting of thermal power load factor, and the weighting of wind power / photovoltaic power output during specific periods) and the graded early warning thresholds (such as relaxing the early warning threshold for short-term fluctuations in new energy and tightening the threshold for stable operation periods of thermal power); and corresponding refined management and control suggestions can be output simultaneously. In practical applications, the aforementioned control recommendations can be directly integrated with the system's planning decomposition module, early warning configuration module, and unified reporting portal to achieve a closed-loop control process of "analysis-optimization-implementation-feedback," continuously improving the accuracy of planned electricity consumption control and cross-departmental collaboration efficiency.

[0037] Corresponding to the planned power consumption control method based on multi-indicator collaborative tracking described in the above embodiments, Figure 2 This diagram illustrates the structural block diagram of a planned power consumption management system based on multi-indicator collaborative tracking, as provided in an embodiment of this application. For ease of explanation, only the parts relevant to the embodiment of this application are shown. The system includes:

[0038] The indicator system construction module is used to construct a multi-indicator collaborative system; the multi-indicator collaborative system includes, but is not limited to, planning indicators, execution indicators, forecasting indicators, and deviation indicators. The data acquisition and standardization module is used for data acquisition and standardization processing; the acquired data includes the core data sources of planning, execution, and forecasting indicators in the multi-indicator collaborative system, including real-time operation data, planning and transaction-related data, site reporting data, and cross-system related data; The collaborative tracking module is used for multi-indicator collaborative tracking. It triggers tiered early warnings by comparing execution, planning and forecast indicators in real time, automatically generates supervision orders for deviation events, and tracks supervision feedback until the deviation is closed. The analysis and optimization module is used for multi-dimensional analysis and optimization; based on the analysis results, it optimizes the plan decomposition rules and hierarchical early warning thresholds, and outputs corresponding control recommendations.

[0039] The planned indicators include at least the group's annual planned electricity volume, the company's annual or monthly planned electricity volume, and the annual or monthly reported planned electricity volume of the power stations; The execution-related indicators include at least the daily or monthly power generation of the power station, cumulative power generation, and plan completion rate; The forecasting indicators include at least forecasted electricity volume based on historical averages and forecasted electricity volume based on power generation type characteristics. The deviation indicators include at least the deviation of the plan completion rate and the deviation of the forecast.

[0040] The predicted power generation based on power generation type characteristics includes at least one of the following: hydropower characteristic-corrected predicted power generation, thermal power characteristic-corrected predicted power generation, wind power characteristic-corrected predicted power generation, and photovoltaic characteristic-corrected predicted power generation.

[0041] The multidimensional analysis includes: Year-on-year analysis: By annual, monthly, and daily granularity, the performance indicators of the same period this year are compared with the data of the same period in the past N years to quantify the historical operating trend; Planned comparison analysis: Based on the three-level management and control dimensions of group-company-station, the actual cumulative power generation at each level is compared with the corresponding planned indicators for the period to pinpoint the level at which the plan execution deviation is attributed and the weak links in coordination. Prediction deviation analysis: Classified by hydropower, thermal power, wind power, and photovoltaic power, calculate the deviation value and deviation rate between the predicted power generation and the actual power generation, evaluate the accuracy of different prediction logics, and quantitatively verify the correction effect of different power generation type characteristics.

[0042] Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by program instructions and related hardware. The program can be stored in a computer-readable storage medium, such as ROM, RAM, disk, optical disk, etc.

[0043] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A planned power consumption control method based on multi-indicator collaborative tracking, characterized in that, The method includes: Construct a multi-indicator collaborative system; the multi-indicator collaborative system includes, but is not limited to, planning indicators, execution indicators, forecasting indicators, and deviation indicators; Data collection and standardization processing are performed; the collected data includes the core data sources of planning, execution, and forecasting indicators in the multi-indicator collaborative system, including real-time operation data, planning and transaction-related data, site reporting data, and cross-system related data; Multi-indicator collaborative tracking is implemented, and hierarchical early warnings are triggered by real-time comparison of execution, planning and forecast indicators. Supervision orders are automatically generated for deviation events, and supervision feedback is tracked until the deviation is closed. Conduct multi-dimensional analysis and optimization; based on the analysis results, optimize the plan decomposition rules and hierarchical early warning thresholds, and output corresponding control recommendations.

2. The planned power consumption management method based on multi-indicator collaborative tracking as described in claim 1, characterized in that, The planned indicators include at least the group's annual planned electricity volume, the company's annual or monthly planned electricity volume, and the annual or monthly reported planned electricity volume of the power stations; The execution-related indicators include at least the daily or monthly power generation of the power station, cumulative power generation, and plan completion rate; The forecasting indicators include at least forecasted electricity volume based on historical averages and forecasted electricity volume based on power generation type characteristics. The deviation indicators include at least the deviation of the plan completion rate and the deviation of the forecast.

3. The planned power consumption management method based on multi-indicator collaborative tracking as described in claim 2, characterized in that, The predicted power generation based on power generation type characteristics includes at least one of the following: hydropower characteristic-corrected predicted power generation, thermal power characteristic-corrected predicted power generation, wind power characteristic-corrected predicted power generation, and photovoltaic characteristic-corrected predicted power generation.

4. The planned power consumption control method based on multi-indicator collaborative tracking as described in claim 3, characterized in that, The calculation logic for the predicted power generation with hydropower characteristic correction is as follows: average power generation in the same period of the past N years × water inflow deviation coefficient; wherein, the water inflow deviation coefficient is the ratio of the actual water inflow in the current month to the average water inflow in the same period of the past N years; The calculation logic for the predicted power generation with thermal power characteristic correction is as follows: average power generation in the same period of the past N years × load factor correction coefficient; wherein, the load factor correction coefficient is the ratio of the predicted load factor of the power grid in the current month to the average load factor in the same period of the past N years, and can be superimposed with the fuel supply correction coefficient according to actual demand; The calculation logic for wind power characteristic correction prediction is as follows: average power generation in the same period of the past N years × wind speed correction coefficient; wherein, the wind speed correction coefficient is the ratio of the actual average wind speed of the current month to the average wind speed in the same period of the past N years, and the predicted value for the time period can be further optimized according to the daily time-sharing wind speed distribution. The calculation logic for photovoltaic characteristic correction prediction power is as follows: average power generation in the same period of the past N years × (irradiance correction coefficient × sunshine duration correction coefficient); where the irradiance correction coefficient is the ratio of the actual average irradiance intensity of the current month to the average irradiance intensity of the same period of the past N years; and the sunshine duration correction coefficient is the ratio of the actual sunshine duration of the current month to the average sunshine duration of the same period of the past N years.

5. The planned power consumption management method based on multi-indicator collaborative tracking as described in claim 1, characterized in that, The steps for multi-indicator collaborative tracking, triggering tiered early warnings by real-time comparison of execution, planning, and forecasting indicators, automatically generating follow-up orders for deviation events, and tracking follow-up feedback until the deviation is closed include: Based on preset electricity decomposition rules, the planned electricity consumption of the group or company is automatically decomposed into annual monthly, monthly daily, and daily hourly electricity consumption. Real-time comparison of execution indicators with planning indicators and forecast indicators; when deviation indicators exceed preset thresholds, a tiered warning is triggered. For deviation events that trigger tiered warnings, a supervision order is automatically generated to identify the responsible unit, responsible person, and completion deadline, and to track and supervise feedback until the deviation is closed.

6. The planned power consumption control method based on multi-indicator collaborative tracking as described in claim 5, characterized in that, The power decomposition rules include: configuring decomposition weights according to power generation type; configuring monthly weights for hydropower based on the proportion of average monthly power generation over the past N years; configuring daily time-of-use weights for thermal power based on historical load factor curves; and configuring time-of-use weights for wind power and photovoltaic power based on historical output curves. If a power station does not have its own decomposition rules, it can automatically inherit the general decomposition rules for its power generation type. N is selected from 3 to 10 years based on the power station's data accumulation.

7. The planned power consumption control method based on multi-indicator collaborative tracking as described in claim 5, characterized in that, The method for comparing execution indicators with planning indicators and forecasting indicators is as follows: First, according to the three-level control dimension of group-company-site, the execution indicators of the same organizational level are associated with the planning indicators of the corresponding period, and the comparison between the two is achieved by calculating the plan completion rate; Then, the execution-related indicators and prediction-related indicators at the same organizational level and time granularity are associated, and the prediction deviation and prediction deviation rate are calculated to compare the two and quantify the degree of fit between the actual execution and the prediction value. The comparison results are synchronously mapped to the corresponding deviation indicators, providing direct data support for subsequent determination of whether the preset threshold is exceeded and triggering graded early warning.

8. The planned power consumption management method based on multi-indicator collaborative tracking as described in claim 1, characterized in that, The supervision order incorporates a three-tiered responsibility traceability mechanism at the group, company, and power plant levels. It automatically matches the responsible unit and extracts the responsible person based on the organizational level to which the deviation event belongs and the power generation type. At the same time, it dynamically sets the completion deadline based on the severity of the deviation.

9. The planned power consumption management method based on multi-indicator collaborative tracking as described in claim 1, characterized in that, The multidimensional analysis includes: Year-on-year analysis: By annual, monthly, and daily granularity, the performance indicators of the same period this year are compared with the data of the same period in the past N years to quantify the historical operating trend; Planned comparison analysis: Based on the three-level management and control dimensions of group-company-station, the actual cumulative power generation at each level is compared with the corresponding planned indicators for the period to pinpoint the level at which the plan execution deviation is attributed and the weak links in coordination. Prediction deviation analysis: Classified by hydropower, thermal power, wind power, and photovoltaic power, calculate the deviation value and deviation rate between the predicted power generation and the actual power generation, evaluate the accuracy of different prediction logics, and quantitatively verify the correction effect of different power generation type characteristics.

10. A planned power consumption management system based on multi-indicator collaborative tracking, characterized in that, The system includes: The indicator system construction module is used to construct a multi-indicator collaborative system; the multi-indicator collaborative system includes, but is not limited to, planning indicators, execution indicators, forecasting indicators, and deviation indicators. The data acquisition and standardization module is used for data acquisition and standardization processing; the acquired data includes the core data sources of planning, execution, and forecasting indicators in the multi-indicator collaborative system, including real-time operation data, planning and transaction-related data, site reporting data, and cross-system related data; The collaborative tracking module is used for multi-indicator collaborative tracking. It triggers tiered early warnings by comparing execution, planning and forecast indicators in real time, automatically generates supervision orders for deviation events, and tracks supervision feedback until the deviation is closed. The analysis and optimization module is used for multi-dimensional analysis and optimization; based on the analysis results, it optimizes the plan decomposition rules and hierarchical early warning thresholds, and outputs corresponding control recommendations.

Images