Energy efficiency monitoring analysis system based on big data
By constructing an energy consumption characteristic disturbance resistance metric and a dual-condition mapping mode, the problems of large error and poor reliability of the energy efficiency monitoring and analysis system are solved, and high accuracy and high reliability analysis of energy efficiency monitoring are achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-25
- Publication Date
- 2026-04-10
AI Technical Summary
Existing energy efficiency monitoring and analysis systems suffer from problems such as improper handling of extreme values, neglect of differences in energy consumption span, and failure to address the uneven distribution of energy consumption, resulting in large monitoring and analysis errors. Furthermore, their systems suffer from coarse division of operating conditions and failure to quantify the uncertainty of energy consumption status, leading to poor reliability of monitoring and analysis.
We construct a triple-mechanism energy consumption characteristic disturbance resistance measure that integrates span calibration, distributed balance correction and exponential compression, define a piecewise benchmark guiding function and benchmark cross random factor, design a dedicated mapping mode for steady state and dynamic dual operating conditions, and construct a triple calculation system for load rate fluctuation coefficient to achieve accurate analysis of energy efficiency monitoring data.
Significantly improves the accuracy and reliability of energy efficiency monitoring and analysis, accurately resists extreme value interference, adapts to energy consumption characteristics in multiple scenarios, fits the energy consumption characteristics of equipment throughout its entire life cycle, and improves the adaptability and reliability of analysis.
Smart Images
Figure CN121834306A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of energy efficiency monitoring, specifically to an energy efficiency monitoring and analysis system based on big data. Background Technology
[0002] An energy efficiency monitoring and analysis system refers to a technical system that collects energy consumption data from equipment and scenarios through sensors and other devices, and then processes, analyzes, calculates, assesses the status of the data, and identifies anomalies to achieve real-time monitoring of energy consumption status. However, general energy efficiency monitoring and analysis systems suffer from problems such as improper handling of extreme values, neglecting differences in energy consumption spans, and failing to address the issue of uneven energy consumption distribution. Furthermore, they often suffer from coarse division of operating conditions, failure to quantify uncertainties in energy consumption status, and weak ability to capture energy consumption drift caused by equipment aging and process adjustments, leading to poor reliability in monitoring and analysis. Summary of the Invention
[0003] To address the aforementioned issues and overcome the shortcomings of existing technologies, this invention provides a big data-based energy efficiency monitoring and analysis system. Addressing the problems of improper extreme value handling, neglecting differences in energy consumption spans, and failing to resolve uneven energy consumption distribution in general energy efficiency monitoring and analysis systems, which lead to large monitoring and analysis errors, this solution constructs a triple-mechanism energy consumption characteristic disturbance resistance measure that integrates span calibration, distribution balance correction, and exponential compression. This accurately resists extreme value interference and adapts to energy consumption characteristics across multiple scenarios. By defining a piecewise benchmark guiding function and designing an energy efficiency-adaptive benchmark cross-random factor, it ensures that the benchmark search direction is accurate, controllable, and highly relevant. By accurately reflecting actual energy consumption characteristics, this solution significantly improves the accuracy of energy efficiency monitoring and analysis. Addressing the issues of coarse operating condition classification, unquantified energy consumption uncertainty, and weak ability to capture energy consumption drift caused by equipment aging and process adjustments—leading to poor reliability in general energy efficiency monitoring and analysis systems—this solution constructs a triple calculation system for load rate fluctuation coefficients, aligning with the steady-state operating load standards of industrial equipment and adapting to the operating characteristics of different equipment types. It also designs a dedicated mapping mode for both steady-state and dynamic operating conditions, aligning with the full-cycle energy consumption characteristics of equipment and fully covering operating scenarios. This enhances the reliability and adaptability of energy efficiency monitoring and analysis.
[0004] The technical solution adopted by the present invention is as follows: The present invention provides an energy efficiency monitoring and analysis system based on big data, including a data acquisition module, an energy efficiency monitoring data preprocessing module, an operating condition energy efficiency benchmark generation module, an energy efficiency status quantification module, a collaborative iterative optimization module, and an energy efficiency monitoring and analysis module;
[0005] The data acquisition module acquires energy efficiency monitoring data and constructs an energy consumption dataset.
[0006] The energy efficiency monitoring data preprocessing module constructs an energy consumption characteristic anti-disturbance metric based on span calibration.
[0007] The operating condition energy efficiency benchmark generation module iteratively selects the optimal initial benchmark based on the benchmark guiding function.
[0008] The energy efficiency status quantification module determines the operating conditions based on load rate fluctuations and designs mapping modes to realize energy efficiency status quantification.
[0009] The collaborative iterative optimization module dynamically updates the energy efficiency benchmark and allocates energy efficiency monitoring data to obtain the clustering results of the energy efficiency monitoring data.
[0010] The energy efficiency monitoring and analysis module performs energy efficiency monitoring and analysis based on the clustering results of energy efficiency monitoring data.
[0011] Furthermore, the data acquisition module acquires energy efficiency monitoring data, including core energy consumption indicators and operating condition related indicators; and performs data preprocessing to obtain an energy consumption dataset.
[0012] Furthermore, the energy efficiency monitoring data preprocessing module combines span calibration, distribution balance correction, and exponential compression mechanisms to construct an energy consumption characteristic disturbance resistance metric.
[0013] Furthermore, the operating condition energy efficiency benchmark generation module specifically includes:
[0014] Define a benchmark guiding function; used to control the benchmark search direction;
[0015] Define a benchmark crossover random factor to fit the energy efficiency characteristic distribution; first, generate energy consumption matching random numbers, then calculate the benchmark crossover offset; and finally obtain the derived energy efficiency benchmark.
[0016] Furthermore, the energy efficiency status quantification module specifically includes:
[0017] The operating condition is initially determined, and the evaluation load rate fluctuation coefficient of the energy efficiency benchmark is calculated. The load rate fluctuation coefficient of the current energy efficiency monitoring sample is also calculated. The evaluation load rate fluctuation coefficient is constructed. The operating condition type is initially determined based on the evaluation load rate fluctuation coefficient.
[0018] Steady-state operating condition mapping mode definition; firstly, construct the energy efficiency state smoothing mapping function; then obtain the corrected energy efficiency fit, non-fit, and hesitation.
[0019] Define the dynamic operating condition mapping mode; calculate the corrected energy efficiency fit, non-fit, and hesitation degree of the dynamic operating condition.
[0020] Furthermore, the collaborative iterative optimization module specifically includes:
[0021] Initial energy efficiency baseline generation;
[0022] Define a baseline optimization fitness function with the goal of optimal energy consumption, and integrate disturbance resistance metric and fit.
[0023] Energy efficiency standards are updated dynamically;
[0024] Energy efficiency fit matrix update;
[0025] Design iterative decision conditions; obtain iterative results for the energy consumption dataset.
[0026] Furthermore, the energy efficiency monitoring and analysis module identifies abnormal energy consumption based on the iterative results of the energy consumption dataset, thereby achieving energy efficiency monitoring and analysis.
[0027] The beneficial effects achieved by the present invention using the above solution are as follows:
[0028] (1) In view of the problems that general energy efficiency monitoring and analysis systems have, such as improper handling of extreme values, neglect of differences in energy consumption span, and failure to solve the problem of uneven energy consumption distribution, which leads to large monitoring and analysis errors, this solution constructs a triple mechanism of energy consumption characteristic anti-disturbance measurement that integrates span calibration, distribution balance correction and exponential compression, so as to accurately resist extreme value interference and adapt to the energy consumption characteristics of multiple scenarios; by defining a segmented benchmark guiding function and designing a benchmark cross random factor for energy efficiency adaptation, the benchmark search direction is ensured to be accurate and controllable and highly consistent with the actual energy consumption characteristic distribution, which ultimately significantly improves the accuracy of energy efficiency monitoring and analysis.
[0029] (2) In view of the problems of general energy efficiency monitoring and analysis systems, such as rough division of working conditions, lack of quantification of energy consumption status uncertainty, weak ability to capture energy consumption drift caused by equipment aging and process adjustment, resulting in poor monitoring and analysis reliability, this solution constructs a triple calculation system of load rate fluctuation coefficient, which conforms to the steady-state operating load standard of industrial equipment and adapts to the working conditions of different types of equipment; designs a dedicated mapping mode for steady-state and dynamic dual working conditions, which conforms to the energy consumption characteristics of the equipment throughout the entire cycle and fully covers the operating scenarios; thereby improving the reliability and adaptability of energy efficiency monitoring and analysis. Attached Figure Description
[0030] Figure 1 This is a flowchart illustrating an energy efficiency monitoring and analysis system based on big data, as provided by the present invention.
[0031] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. Detailed Implementation
[0032] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.
[0033] In the description of this invention, it should be understood that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.
[0034] Example 1, see Figure 1 The present invention provides an energy efficiency monitoring and analysis system based on big data, including a data acquisition module, an energy efficiency monitoring data preprocessing module, an operating condition energy efficiency benchmark generation module, an energy efficiency status quantification module, a collaborative iterative optimization module, and an energy efficiency monitoring and analysis module;
[0035] The data acquisition module acquires energy efficiency monitoring data, constructs an energy consumption dataset, and sends the data to the energy efficiency monitoring data preprocessing module.
[0036] The energy efficiency monitoring data preprocessing module constructs an energy consumption characteristic anti-disturbance metric based on span calibration and sends the data to the operating condition energy efficiency benchmark generation module.
[0037] The operating condition energy efficiency benchmark generation module iteratively selects the optimal initial benchmark based on the benchmark guiding function, and sends the data to the energy efficiency status quantification module.
[0038] The energy efficiency status quantification module determines the operating conditions based on load rate fluctuations, designs mapping modes for each, and realizes energy efficiency status quantification; and sends the data to the collaborative iterative optimization module.
[0039] The collaborative iterative optimization module dynamically updates the energy efficiency benchmark, allocates energy efficiency monitoring data, obtains the clustering results of the energy efficiency monitoring data, and sends the data to the energy efficiency monitoring and analysis module.
[0040] The energy efficiency monitoring and analysis module performs energy efficiency monitoring and analysis based on the clustering results of energy efficiency monitoring data.
[0041] Example 2, see Figure 1 This embodiment is based on the above embodiment. The data acquisition module acquires energy efficiency monitoring data and collects core energy consumption indicators and operating condition related indicators by deploying a distributed sensor network. The core energy consumption indicators include equipment power, cumulative energy consumption, and operating time. The operating condition related indicators include ambient temperature, load rate, voltage, and current. Data preprocessing is performed, including missing value interpolation (linear interpolation method), redundant data removal (variance threshold method), and data format standardization to obtain an energy consumption dataset.
[0042] Example 3, see Figure 1This embodiment is based on the above embodiment. The energy efficiency monitoring data preprocessing module addresses the fact that even after preliminary preprocessing, the data still contains extreme values caused by sensor momentary faults and voltage fluctuations. Therefore, it incorporates span calibration. Distribution balance correction With exponential compression mechanism Construct an energy consumption characteristic disturbance immunity metric, expressed as: ; ;in, This is the p-th energy efficiency monitoring sample; It is the qth energy efficiency benchmark center. yes The j-th energy consumption characteristic value; It is the overall variance of energy consumption data, adapted to the energy consumption fluctuation characteristics of different industries, and controls the distance decay rate; It is the energy consumption characteristic span calibration coefficient; It is the energy consumption distribution balance coefficient, which corrects the energy consumption deviation during peak and valley periods and reduces the weight of extreme values on the measurement results. Its value is (0,2]. yes The j-th energy consumption characteristic; It is the working condition adaptation factor, with a value of [0,1].
[0043] Example 4, see Figure 1 This embodiment is based on the above embodiment. The operating condition energy efficiency benchmark generation module constructs a benchmark guiding function and iteratively filters the optimal initial benchmark that fits the actual energy consumption distribution. The specific operation is as follows:
[0044] Define a benchmark guiding function; used to control the benchmark search direction and avoid blind searching. Represented as: ;in, It is the qth feature of the derived energy efficiency benchmark; It is the mean of the parental baseline characteristics, guiding the derivatives to move towards the historical best energy efficiency; It is the variance of the parental baseline characteristics, which controls the baseline adjustment range and ensures stability;
[0045] Define a baseline crossover random factor to fit the energy efficiency characteristic distribution; first, generate energy consumption-adaptive random numbers. Adapting to the energy efficiency characteristic distribution, expressed as: Then calculate the reference cross offset. Used for fine-tuning the reference amplitude, expressed as: This leads to the generation of a derived energy efficiency benchmark, expressed as: ;in, and These are two parental energy efficiency benchmark values; and These are two derived energy efficiency benchmark values; It is the j-th eigenvalue of the best parental energy efficiency benchmark in the parental energy efficiency benchmark set in the current iteration round; and It is the baseline cross offset between the two parents.
[0046] By performing the above operations, this solution addresses the problems of improper handling of extreme values, neglect of differences in energy consumption span, and failure to resolve the uneven distribution of energy consumption in general energy efficiency monitoring and analysis systems, which lead to large monitoring and analysis errors. It constructs a triple-mechanism energy consumption characteristic disturbance resistance metric that integrates span calibration, distribution balance correction, and exponential compression. This metric accurately resists extreme value interference and adapts to energy consumption characteristics in multiple scenarios. By defining a piecewise benchmark guiding function and designing a benchmark cross-random factor for energy efficiency adaptation, it ensures that the benchmark search direction is accurate and controllable and highly consistent with the actual energy consumption characteristic distribution, ultimately significantly improving the accuracy of energy efficiency monitoring and analysis.
[0047] Example 5, see Figure 1 This embodiment is based on the above embodiment. The energy efficiency status quantification module is used when the equipment is often in a transitional operating condition (start-up phase, half-load operation). By quantifying the energy consumption status under different operating conditions, and combining steady-state and dynamic operating conditions to design a dual mapping mode, it describes the deterministic attribution of the operating condition. The specific operation is as follows:
[0048] Based on the initial assessment of operating conditions, the evaluation load rate fluctuation coefficient of the energy efficiency benchmark is calculated and expressed as: ; ; And calculate the load rate fluctuation coefficient of the current energy efficiency monitoring sample, expressed as: ; ; Construct an evaluation load rate fluctuation coefficient , is represented as: ;in, It is the baseline load rate fluctuation coefficient, which reflects the relative fluctuation of the load rate under the standard operating conditions corresponding to the qth energy efficiency baseline. The smaller the value, the more stable the operating conditions corresponding to the baseline. and These are the mean and standard deviation of the baseline load factor, respectively. is the load rate of the i-th energy efficiency monitoring sample in the q-th energy efficiency benchmark; n is the total number of energy efficiency monitoring samples in the benchmark, and i is the index of the energy efficiency monitoring sample in the benchmark; It is the load rate fluctuation coefficient of the current energy efficiency monitoring sample, reflecting the relative fluctuation of the load rate of the p-th energy efficiency monitoring sample to be judged; and These are the mean and standard deviation of the current energy efficiency monitoring sample load rate, respectively. It is the load rate of the p-th energy efficiency monitoring sample during the u-th sampling; N is the time window; This is the minimum value for operating condition evaluation, ranging from [0.01, 0.05]. The operating condition type is initially determined based on the evaluation load rate fluctuation coefficient. If the condition is stable, then it is a steady-state operating condition; otherwise, it is a dynamic operating condition.
[0049] The steady-state operating condition mapping mode is defined, corresponding to the normal load operation scenario of the equipment; firstly, an energy efficiency state smoothing mapping function is constructed. , is represented as: The corrected energy efficiency fit, which aligns with the steady-state attribution, is expressed as: The degree of mismatch and the degree of hesitation (quantifying steady-state uncertainty) are respectively expressed as: ; ;in, It is the initial energy efficiency fit, which represents the initial fit degree of the p-th energy efficiency monitoring sample to the q-th energy efficiency benchmark; It is the steady-state adjustment coefficient, with values of (0.5, 3); It is the energy efficiency fit after steady-state correction; and These are steady-state non-fitness and steady-state hesitation, respectively.
[0050] The dynamic operating condition mapping mode is defined to correspond to equipment start-up / shutdown and load change scenarios; the corrected energy efficiency fit is expressed as: The degree of non-fitness and the degree of hesitation are respectively represented as: ; ;in, , and These are, respectively, the energy efficiency fit after dynamic correction, the dynamic misfit, and the dynamic hesitation. It is the dynamic operating condition adjustment coefficient, with a value range of (1.2, 5.0);
[0051] Breaking away from black-and-white classifications, it accurately depicts the energy efficiency status of transitional operating conditions; hesitation, as an indicator of energy consumption stability, provides additional reference for equipment operation and maintenance.
[0052] Example 6, see Figure 1 The collaborative iterative optimization module addresses energy consumption drift during equipment operation (equipment aging, process adjustments). Therefore, a closed-loop architecture of benchmark optimization—state quantification—iterative correction is constructed to dynamically update the benchmark and state quantification results, ensuring that the analysis results accurately reflect real-time energy consumption changes. The specific operation is as follows:
[0053] The initial energy efficiency baseline is generated as follows: ;in, It is the qth initial energy efficiency benchmark; s is the total number of energy efficiency monitoring samples; It is the total number of energy efficiency monitoring samples corresponding to the qth benchmark;
[0054] Define the baseline optimization fitness function, with energy optimization as the objective, as follows: ; It is the energy efficiency benchmark optimization adaptation constant, with a value of [10, 100]. The energy efficiency clustering objective function, which integrates disturbance resistance metric and fit, is expressed as: Where fitness is the baseline optimization fitness function; q is the q-th energy efficiency objective function; M is the total number of energy efficiency benchmarks; m is the total number of energy efficiency monitoring samples during the iteration process; It is the energy efficiency fit after the general-purpose modification. is a general working condition placeholder, with a value set of {s,d}; k is the weighting index, with a value of [1.5,2.5]; This is the updated q-th energy efficiency benchmark;
[0055] The energy efficiency benchmark is dynamically updated to match real-time energy consumption distribution, and is represented as follows: ;
[0056] The energy efficiency fit matrix is updated, and the accuracy is improved by combining disturbance rejection metrics, as follows: ;in, It is a general-purpose fit matrix element, which is the corrected fit of the p-th energy efficiency monitoring sample to the q-th benchmark; This is the updated j-th energy efficiency benchmark; the energy efficiency monitoring samples are assigned to the benchmark with the highest fit.
[0057] Iterative decision condition design, if If the iteration stops, then the iteration stops; where, It is the fitness value of the current iteration round; It is the fitness value from the previous iteration round; It is the iterative convergence threshold, with a value range of
[10] . -5 10 -3 If the maximum number of iterations is reached, the parameters are adjusted and the clustering is re-established based on the particle swarm search algorithm.
[0058] By performing the above operations, this solution addresses the problems of general energy efficiency monitoring and analysis systems, such as coarse division of operating conditions, lack of quantification of energy consumption uncertainty, and weak ability to capture energy consumption drift caused by equipment aging and process adjustments, leading to poor monitoring and analysis reliability. This solution constructs a triple calculation system for load rate fluctuation coefficients, conforming to the steady-state operating load standards of industrial equipment and adapting to the operating characteristics of different types of equipment. It also designs a dedicated mapping mode for both steady-state and dynamic operating conditions, conforming to the full-cycle energy consumption characteristics of equipment and fully covering operating scenarios. This improves the reliability and adaptability of energy efficiency monitoring and analysis.
[0059] Example 7, see Figure 1 This embodiment is based on the above embodiment. The energy efficiency monitoring and analysis module identifies abnormal energy consumption based on the iterative results of the energy consumption dataset, thereby realizing energy efficiency monitoring and analysis; based on the dual threshold determination of anti-disturbance distance and hesitation degree, if... If so, it is determined to be abnormal, its location is determined, and it is reported to the management personnel; It is the anomaly threshold for anti-interference distance, with a value of (0, 0.8); It is the threshold for abnormal hesitation, with a value of (0, 0.3); yes The corresponding energy efficiency benchmark; It is the degree of hesitation of the s-th energy efficiency monitoring sample belonging to the q-th energy consumption benchmark.
[0060] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention.
[0061] The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.
Claims
1. A big data based energy efficiency monitoring and analysis system, characterized in that: The system comprises a data acquisition module, an energy efficiency monitoring data preprocessing module, an operating condition energy efficiency benchmark generation module, an energy efficiency state quantification module, a collaborative iterative optimization module, and an energy efficiency monitoring analysis module. The data acquisition module acquires energy efficiency monitoring data and constructs an energy consumption dataset. The energy efficiency monitoring data preprocessing module constructs an energy consumption feature anti-interference metric based on span calibration. The operating condition energy efficiency benchmark generation module iteratively selects an optimal initial benchmark based on a benchmark guide function. The energy efficiency state quantification module determines operating conditions based on load rate fluctuation, respectively designs mapping modes, and realizes energy efficiency state quantification. The collaborative iterative optimization module dynamically updates energy efficiency benchmarks and allocates energy efficiency monitoring data to obtain clustering results of energy efficiency monitoring data. The energy efficiency monitoring analysis module realizes energy efficiency monitoring analysis based on the clustering results of energy efficiency monitoring data.
2. The energy efficiency monitoring and analysis system based on big data according to claim 1, characterized in that: The energy efficiency monitoring data preprocessing module combines span calibration, distribution balance correction, and exponential compression mechanism to construct an energy consumption feature anti-interference metric.
3. The energy efficiency monitoring and analysis system based on big data according to claim 2, characterized in that: The operating condition energy efficiency benchmark generation module specifically comprises: Defining a benchmark guide function for controlling the benchmark search direction. Defining a benchmark cross random factor to adapt to energy efficiency feature distribution; first generating an energy consumption adaptive random number, then calculating a benchmark cross offset; and then deriving energy efficiency benchmark generation.
4. The energy efficiency monitoring and analysis system based on big data according to claim 3, characterized in that: The energy efficiency state quantification module specifically comprises: Operating condition preliminary determination; Steady-state operating condition mapping mode definition; first constructing an energy efficiency state smoothing mapping function; obtaining corrected energy efficiency fitness, non-fitness, and hesitation degree; Dynamic operating condition mapping mode definition; calculating the corrected energy efficiency fitness, non-fitness, and hesitation degree of dynamic operating conditions.
5. The big data based energy efficiency monitoring and analysis system as claimed in claim 4, wherein: The operating condition preliminary determination is to calculate the evaluation load rate fluctuation coefficient of the energy efficiency benchmark; and calculate the load rate fluctuation coefficient of the current energy efficiency monitoring sample; construct the evaluation load rate fluctuation coefficient; and preliminarily determine the operating condition type based on the evaluation load rate fluctuation coefficient.
6. The energy efficiency monitoring and analysis system based on big data according to claim 5, characterized in that: The collaborative iterative optimization module specifically comprises: Initial energy efficiency benchmark generation; Defining a benchmark optimization fitness function to optimize energy consumption as the target, integrating anti-interference metric and fitness; Energy efficiency benchmark dynamic update; Energy efficiency fitness matrix update; Iterative determination condition design; obtaining the iterative results of the energy consumption dataset.
7. The big data based energy efficiency monitoring and analysis system of claim 6, wherein: The data acquisition module acquires energy efficiency monitoring data, including core energy consumption indicators and operating condition associated indicators; and performs data preprocessing; obtaining an energy consumption dataset.
8. The energy efficiency monitoring and analysis system based on big data according to claim 7, characterized in that: The energy efficiency monitoring analysis module is based on the iterative results of the energy consumption dataset to identify abnormal energy consumption, and then realizes energy efficiency monitoring analysis.