CN118071168A - Comprehensive energy management system - Google Patents
Comprehensive energy management system Download PDFInfo
- Publication number
- CN118071168A CN118071168A CN202410131713.5A CN202410131713A CN118071168A CN 118071168 A CN118071168 A CN 118071168A CN 202410131713 A CN202410131713 A CN 202410131713A CN 118071168 A CN118071168 A CN 118071168A
- Authority
- CN
- China
- Prior art keywords
- prediction
- hidden danger
- predicted
- abnormal
- danger assessment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 230000002159 abnormal effect Effects 0.000 claims abstract description 83
- 238000000034 method Methods 0.000 claims abstract description 61
- 238000007405 data analysis Methods 0.000 claims abstract description 33
- 238000012545 processing Methods 0.000 claims abstract description 25
- 238000004458 analytical method Methods 0.000 claims abstract description 17
- 238000012544 monitoring process Methods 0.000 claims abstract description 16
- 238000007726 management method Methods 0.000 claims description 18
- 238000004364 calculation method Methods 0.000 claims description 14
- 230000005856 abnormality Effects 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 7
- 238000011156 evaluation Methods 0.000 claims description 4
- NFEIBWMZVIVJLQ-UHFFFAOYSA-N mexiletine hydrochloride Chemical compound [Cl-].CC([NH3+])COC1=C(C)C=CC=C1C NFEIBWMZVIVJLQ-UHFFFAOYSA-N 0.000 claims description 3
- 239000002699 waste material Substances 0.000 abstract description 3
- 238000004590 computer program Methods 0.000 description 4
- 238000005265 energy consumption Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 238000005381 potential energy Methods 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000013210 evaluation model Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- 238000013479 data entry Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000006866 deterioration Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000002737 fuel gas Substances 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012821 model calculation Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Health & Medical Sciences (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a comprehensive energy management system, which particularly relates to the technical field of energy management, and comprises a prediction model monitoring module, a central processing module, an analysis module and a comparison module, wherein the modules are connected through signals, the prediction process of an energy prediction model is monitored, a plurality of characteristic data related to the prediction process of the energy prediction model are extracted from monitored data, a prediction hidden danger assessment index is generated, the prediction process of the energy prediction model is assessed, the prediction process with larger hidden danger is marked, a data analysis set is established according to a marking result, the data analysis set is analyzed, a repair plan of the prediction model is determined, obvious abnormal risks appear in the stability of the energy prediction model, an early warning signal is sent, the stability and the accuracy of prediction are improved, the energy waste and decision errors caused by the instability of the model are avoided, the energy saving potential is identified, and the energy utilization efficiency is improved.
Description
Technical Field
The invention relates to the technical field of energy management, in particular to an integrated energy management system.
Background
The comprehensive energy management system adopts a layered distributed system architecture, collects and processes various classified energy consumption data of electric power, fuel gas, water and the like of a building, analyzes the energy consumption condition of the building, realizes energy saving application and the like of the building, uses the data from multiple angles such as energy supply, energy management, load management, energy consumption analysis and prediction, energy operation and maintenance operation and the like, and completes energy management.
In the energy consumption analysis prediction stage, a prediction model is generally used for predicting future energy demands so as to timely adjust and optimize energy supply and distribution, and the prediction result is inaccurate due to potential hazards of the stability of the prediction model in the long-time use process, and whether the inaccuracy exists or not can not be judged accidentally, so that excessive time and resources are spent for diagnosis, repair and remodelling.
In order to solve the above-mentioned defect, a technical scheme is provided.
Disclosure of Invention
In order to overcome the above-mentioned drawbacks of the prior art, an embodiment of the present invention provides an integrated energy management system to solve the above-mentioned problems of the prior art.
In order to achieve the above purpose, the present invention provides the following technical solutions:
the comprehensive energy management system comprises a prediction model monitoring module, a central processing module, an analysis module and a comparison module, wherein the modules are connected through signals;
the prediction model monitoring module is used for monitoring the prediction process of the energy prediction model, preprocessing the monitored data to obtain multiple characteristic data of the prediction process, and transmitting the multiple characteristic data to the central processing module;
The central processing module establishes a data processing model according to the multiple feature data, generates a predicted hidden danger assessment index, marks the current predicted hidden danger assessment index, and transmits the marked predicted hidden danger assessment index to the analysis module;
The analysis module is used for establishing a data analysis set according to the marked predicted hidden danger assessment indexes, analyzing the existing predicted hidden danger assessment indexes in the data analysis set, generating a steady-state index and transmitting the steady-state index to the comparison module;
And the comparison module is used for comparing the steady state index with a preset steady state index threshold value and determining a repair plan of the prediction model.
In a preferred embodiment, the plurality of feature data comprises prediction interval error data, prediction process data, the prediction interval error data comprising a confidence interval deviation coefficient, the prediction process data comprising a predicted load representation coefficient and a katon anomaly time coefficient, the confidence interval deviation coefficient, the predicted load representation coefficient, the katon anomaly time coefficient being labeled ZX, FX, KD, respectively.
In a preferred embodiment, the logic for obtaining the confidence interval deviation factor is as follows:
Calculating a confidence interval expressed by the following expression zx=yu±α×se, wherein ZX represents the confidence interval, i.e., zx= [ ZX 1,ZX2 ], YU represents the predicted value, α represents the confidence coefficient, and SE represents the prediction standard error, wherein Where MSE represents the mean square error, x represents the energy prediction model input value,Representing an average value of input values of the energy prediction model, wherein i is the number of times of prediction of the energy prediction model, and n is a positive integer;
comparing the actual observation value with the confidence interval of the predicted result, and calculating the confidence interval deviation coefficient when the actual observation value exceeds the confidence interval of the predicted result, wherein the expression is as follows Where PC represents the deviation value of the confidence interval of the actual observed value and the predicted result, and SJ represents the actual observed value.
In a preferred embodiment, the acquisition logic for predicting the load performance factor is as follows:
The expression of the calculated load expression value is as follows Wherein YC represents the deviation value between the actual observed value and the predicted value, FU represents different load states, the predicted data quantity can be quantized, and beta represents a constant for adjusting or normalizing the calculation result;
comparing the load representation value with a preset load representation value threshold, marking the load representation value as an abnormal load representation value when the load representation value is larger than or equal to the load representation value threshold, and calculating a predicted load representation coefficient according to the abnormal load representation value Wherein j represents the number of predictions of the occurrence of abnormal load representation values, and m is a positive integer;
The capture logic of the katon anomaly time coefficient is as follows:
calculating the single predicted duty cycle of the break-in time, the calculation expression of which is as follows Where KD represents the katen time of the single prediction process and ZH represents the execution time of the single prediction;
Comparing the duty ratio of the click time with a preset duty ratio threshold of the click time, marking the duty ratio of the click time as an abnormal duty ratio of the click time when the duty ratio of the click time is larger than or equal to the duty ratio threshold of the click time, and calculating a click abnormal time coefficient according to the abnormal duty ratio of the click time, wherein the expression is as follows Wherein AK represents abnormal cartoon time duty ratio times, and ZC represents total times predicted by the energy prediction model.
In a preferred embodiment, a predicted hidden danger assessment index is obtained through formula calculation according to a confidence interval deviation coefficient, a predicted load expression coefficient and a katon abnormal time coefficient;
comparing the predicted hidden danger assessment index with a preset predicted hidden danger assessment index threshold;
If the predicted hidden danger assessment index is greater than or equal to the predicted hidden danger assessment index threshold, marking the generated predicted hidden danger assessment index as an abnormal predicted hidden danger assessment index, and transmitting the abnormal predicted hidden danger assessment index to an analysis module;
if the predicted hidden danger assessment index is smaller than the predicted hidden danger assessment index threshold, no other operation is needed.
In a preferred embodiment, a data analysis set is built according to sequentially received anomaly prediction hidden danger assessment indexes, and is marked asWherein/>The method comprises the steps that sequentially received abnormal prediction hidden danger assessment indexes are represented, v represents sequence numbers of the sequentially received abnormal prediction hidden danger assessment indexes, and u is a positive integer;
Acquiring outlier information and interval time information of existing abnormal prediction hidden danger assessment indexes in a data analysis set, wherein the outlier information comprises abnormal prediction hidden danger assessment index outliers, and the interval time information comprises interval time differences; and respectively marking the outlier and the interval time of the abnormal prediction hidden danger assessment index as QA and NP.
In a preferred embodiment, the calculation expression of the anomaly prediction hidden danger assessment index outlier is as follows The method comprises the steps that an abnormal prediction hidden danger assessment index with a number u in a data analysis set is represented, PZ represents the average value of existing abnormal prediction hidden danger assessment indexes in the data analysis set, and ZB represents the standard deviation of the existing abnormal prediction hidden danger assessment indexes in the data analysis set;
The calculation expression of the interval time difference is as follows np= |t δ-tγ |, where t δ represents the reception time of the abnormality prediction hidden danger assessment index numbered u, and t γ represents the reception time of the abnormality prediction hidden danger assessment index numbered u-1.
In a preferred embodiment, calculating a steady state index according to an outlier of the abnormal prediction hidden danger assessment index and an interval time difference through a formula;
Comparing the steady state index with a preset steady state index threshold;
if the steady state index is greater than or equal to the steady state index threshold, an early warning signal is sent out, and operations such as model repair and retraining are carried out after the prediction of the energy prediction model is finished;
if the steady state index is smaller than the steady state index threshold, the early warning signal is not required to be sent, and the operation such as repairing can be performed on the steady state index periodically.
The invention has the technical effects and advantages that:
1. According to the method, the prediction process of the energy prediction model is monitored, the monitored data are preprocessed, a plurality of characteristic data related to the prediction process of the energy prediction model are extracted, a data processing model is built according to the plurality of characteristic data, a prediction hidden danger assessment index is generated, the prediction process of the energy prediction model is assessed, the prediction process with larger hidden danger is marked, a data analysis set is built according to the marked prediction hidden danger assessment index, the existing prediction hidden danger assessment index in the data analysis set is analyzed, a steady state index is generated, the steady state index is compared with a preset steady state index threshold value, a repair plan of the prediction model is determined, obvious abnormal risks appear in the stability of the energy prediction model, early warning signals are sent out, operations such as model repair and retraining are carried out after the prediction of the energy prediction model, the stability and accuracy of the prediction are improved, energy waste and decision errors caused by instability of the model are avoided, the prediction accuracy is improved, the energy saving potential is recognized, an effective energy saving strategy is formulated, and energy utilization efficiency is improved.
Drawings
For the convenience of those skilled in the art, the present invention will be further described with reference to the accompanying drawings;
fig. 1 is a schematic structural diagram of embodiment 1 of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1
The invention provides a comprehensive energy management system shown in figure 1, which comprises a prediction model monitoring module, a central processing module, an analysis module and a comparison module, wherein the modules are connected through signals;
the prediction model monitoring module is used for monitoring the prediction process of the energy prediction model, preprocessing the monitored data to obtain multiple characteristic data of the prediction process, and transmitting the multiple characteristic data to the central processing module;
The central processing module establishes a data processing model according to the multiple feature data, generates a predicted hidden danger assessment index, marks the current predicted hidden danger assessment index, and transmits the marked predicted hidden danger assessment index to the analysis module;
The analysis module is used for establishing a data analysis set according to the marked predicted hidden danger assessment indexes, analyzing the existing predicted hidden danger assessment indexes in the data analysis set, generating a steady-state index and transmitting the steady-state index to the comparison module;
The comparison module is used for comparing the steady state index with a preset steady state index threshold value and determining a repair plan of the prediction model;
the prediction model monitoring module is used for monitoring the prediction process of the energy prediction model, preprocessing the monitored data to obtain multiple characteristic data of the prediction process, and transmitting the multiple characteristic data to the central processing module;
the prediction model monitoring module comprises a monitoring unit and a preprocessing unit;
the monitoring unit is used for monitoring the prediction process of the energy prediction model and storing the monitored original data;
The preprocessing unit is used for preprocessing the monitored original data and comprises the following steps:
missing value processing, namely detecting and processing missing values in original data, wherein the missing original data can be processed in a filling, deleting or interpolating mode so as to ensure the integrity of the original data.
Outlier processing, detection and processing outliers, outliers may be identified using statistical methods or threshold-based techniques and corrected or deleted.
The data is cleaned, the original data is cleaned, the quality of the original data is ensured, and the method comprises the steps of removing repeated data, correcting wrong data format or data entry errors and the like.
Data conversion, namely performing standardization or normalization processing on original data to eliminate scale differences among different original data, wherein the common data conversion methods comprise Z-score standardization and Min-Max normalization;
extracting characteristics, namely extracting a plurality of characteristic data related to the prediction process of the energy prediction model from original data, wherein the plurality of characteristic data comprise prediction interval error data and prediction process data, the prediction interval error data comprise confidence interval deviation coefficients, and the prediction process data comprise prediction load performance coefficients and cartoon abnormal time coefficients;
The central processing module establishes a data processing model according to the multiple feature data, generates a predicted hidden danger assessment index, marks the current predicted hidden danger assessment index, and transmits the marked predicted hidden danger assessment index to the analysis module;
marking the confidence interval deviation coefficient, the predicted load expression coefficient and the cartoon abnormal time coefficient as ZX, FX and KD respectively;
Establishing a data processing model according to the confidence interval deviation coefficient, the predicted load expression coefficient and the katon abnormal time coefficient to generate a predicted hidden danger assessment index according to the following formula of YP=a 1*ZX+a2*FX+a3 ×KD, wherein YP is the predicted hidden danger assessment index, ZX is the confidence interval deviation coefficient, FX is the predicted load expression coefficient, KD is the katon abnormal time coefficient, and a 1、a2、a3 is the confidence interval deviation coefficient, the predicted load expression coefficient and the proportional coefficient of the katon abnormal time coefficient respectively, wherein a 1>a2>a3 >0;
The confidence interval deviation coefficient is used for measuring the deviation degree of a confidence interval between an actual observed value and a predicted result, and the larger the confidence interval deviation coefficient is, the larger the uncertainty degree of the energy prediction model is, and the higher the potential hazard probability possibly existing in the actual use process of the energy prediction model is; if the actual observed value exceeds the confidence interval, the energy prediction model has high uncertainty on the prediction, which may be caused by that the energy prediction model is not accurate enough to process the data, so that the quality of the data is poor, and the prediction result is affected.
The confidence interval deviation coefficient acquisition logic is as follows:
Calculating a confidence interval expressed as zx=yu±α×se, wherein ZX represents the confidence interval, i.e., zx= [ ZX 1,ZX2 ], YU represents the predicted value, α represents the confidence coefficient, which may be set according to the actual situation, e.g., 95% or 99%, and SE represents the prediction standard error, wherein Where MSE represents mean square error, x represents energy prediction model input value,/>Representing an average value of input values of the energy prediction model, wherein i is the number of times of prediction of the energy prediction model, and n is a positive integer;
comparing the actual observation value with the confidence interval of the predicted result, and calculating the confidence interval deviation coefficient when the actual observation value exceeds the confidence interval of the predicted result, wherein the expression is as follows Wherein PC represents the deviation value of the confidence interval between the actual observed value and the predicted result, and SJ represents the actual observed value;
The predicted load performance coefficient refers to the measurement degree of the prediction accuracy of the energy prediction model under different load states, the higher the predicted load performance coefficient is, the stability of the energy prediction model is possibly affected, the performance of the energy prediction model is reduced, a large amount of or complex model calculation is difficult to process, the prediction accuracy is reduced, when the system load is increased, the model is possibly easier to excessively fit with the existing data, the generalization capability of the model on new data is reduced, and the hidden danger probability of the energy prediction model in the actual use process is increased.
The acquisition logic of the predicted load performance coefficient is as follows:
The expression of the calculated load expression value is as follows Wherein YC represents the deviation value between the actual observed value and the predicted value, FU represents different load states, the predicted data quantity can be quantized, beta represents a constant, and the constant is used for adjusting or standardizing the calculation result and can be set according to the actual situation;
comparing the load representation value with a preset load representation value threshold, marking the load representation value as an abnormal load representation value when the load representation value is larger than or equal to the load representation value threshold, and calculating a predicted load representation coefficient according to the abnormal load representation value Where j represents the number of predictions of the occurrence of abnormal load representation values, and m is a positive integer.
The factor of the abnormal time of the katon is used for measuring the duty ratio degree of the katon time and the single prediction execution time of the prediction process of the energy prediction model, the higher the duty ratio degree is, the more the prediction times of the duty ratio abnormality occur, the stability of the energy prediction model is possibly reduced, the prediction cannot be completed within the expected time, the potential energy saving opportunity and the energy use adjustment opportunity are possibly missed, and if the abnormal duty ratio of the abnormal katon time exists for a long time, the hidden danger probability of the energy prediction model in the actual use process is possibly further increased;
The capture logic of the katon anomaly time coefficient is as follows:
calculating the single predicted duty cycle of the break-in time, the calculation expression of which is as follows Where KD represents the katen time of the single prediction process and ZH represents the execution time of the single prediction;
Comparing the duty ratio of the clamping time with a preset duty ratio threshold value of the clamping time, when the duty ratio of the clamping time is larger than or equal to the duty ratio threshold value of the clamping time, the duty ratio of the clamping time in the prediction process is at an abnormal level and exceeds the acceptable range of the energy prediction model, marking the energy prediction model as the abnormal duty ratio of the clamping time, and when the duty ratio of the clamping time is smaller than the duty ratio threshold value of the clamping time, the duty ratio of the clamping time in the prediction process is at a normal level; calculating a cartoon-in abnormal time coefficient according to the abnormal cartoon-in time proportion, wherein the expression is as follows Wherein AK represents abnormal cartoon time duty ratio times, ZC represents total times predicted by an energy prediction model;
the confidence interval deviation coefficient, the predicted load expression coefficient and the katon abnormal time coefficient are known by the calculation, namely, the prediction hidden danger assessment index is larger, so that the degree of the operation hidden danger existing in the energy prediction model is higher; and conversely, the smaller the confidence interval deviation coefficient, the predicted load performance coefficient and the katon abnormal time coefficient are, namely the smaller the predicted hidden danger assessment index is, the lower the degree of the hidden danger of the operation of the energy prediction model is.
Comparing the predicted hidden danger assessment index with a preset predicted hidden danger assessment index threshold value, assessing the prediction process of the energy prediction model, and marking the prediction process with larger hidden danger;
If the predicted hidden danger assessment index is greater than or equal to the predicted hidden danger assessment index threshold, the hidden danger of the energy prediction model in the prediction process is indicated, the energy prediction model possibly has an unstable risk, the generated predicted hidden danger assessment index is marked as an abnormal predicted hidden danger assessment index, and the abnormal predicted hidden danger assessment index is transmitted to an analysis module;
If the predicted hidden danger assessment index is smaller than the predicted hidden danger assessment index threshold, the energy prediction model does not find the operation hidden danger in the prediction process, and other operations are not needed;
The analysis module is used for establishing a data analysis set according to the marked predicted hidden danger assessment indexes, analyzing the existing predicted hidden danger assessment indexes in the data analysis set, generating a steady-state index and transmitting the steady-state index to the comparison module;
Establishing a data analysis set according to the sequentially received abnormal prediction hidden danger assessment indexes, and marking the data analysis set as Wherein/>The method comprises the steps that sequentially received abnormal prediction hidden danger assessment indexes are represented, v represents sequence numbers of the sequentially received abnormal prediction hidden danger assessment indexes, and u is a positive integer;
acquiring outlier information and interval time information of existing abnormal prediction hidden danger assessment indexes in a data analysis set, wherein the outlier information comprises abnormal prediction hidden danger assessment index outliers, and the interval time information comprises interval time differences; respectively marking outliers and interval time of the abnormal prediction hidden danger assessment indexes as QA and NP;
the outlier of the abnormal prediction hidden danger assessment index is used for measuring the difference degree of the abnormal prediction hidden danger assessment index and the existing abnormal prediction hidden danger assessment index in the data analysis set, wherein the higher the outlier of the abnormal prediction hidden danger assessment index is, the hidden danger existing in the energy prediction model is continuously increased, the instability risk existing in the energy prediction model is greater, the model is repaired immediately, the instability continuous deterioration of the model is avoided, and the potential energy saving opportunity is missed;
The calculation expression of the outlier of the abnormality prediction hidden danger assessment index is as follows The method comprises the steps that an abnormal prediction hidden danger assessment index with a number u in a data analysis set is represented, PZ represents the average value of existing abnormal prediction hidden danger assessment indexes in the data analysis set, and ZB represents the standard deviation of the existing abnormal prediction hidden danger assessment indexes in the data analysis set;
The interval time difference is used for measuring the time difference between two existing abnormal prediction hidden danger assessment indexes in the data analysis set, the larger the interval time difference is, the potential hazards existing in the energy prediction model are possibly accidental, the lower the probability that the energy prediction model itself has instability risks is, the smaller the interval time difference is, and the more compact the situation that the potential hazards are assessed to be abnormal hidden hazards is;
The calculation expression of the interval time difference is as follows np= |t δ-tγ |, where t δ represents the receiving time of the abnormality prediction hidden danger assessment index with the number u, and t γ represents the receiving time of the abnormality prediction hidden danger assessment index with the number u-1;
establishing a steady state evaluation model according to the outlier of the abnormal prediction hidden danger evaluation index and the interval time difference, generating a steady state index, and evaluating the stability of the energy prediction model, wherein the steady state evaluation model is established according to the outlier of the abnormal prediction hidden danger evaluation index and the interval time difference according to the following formula, wherein the formula is as follows, WT=b 1*QA+b2 x NP, wherein the WT represents the steady state index, and b 1、b2 represents the weight factors of the outlier of the abnormal prediction hidden danger evaluation index and the interval time difference respectively, and the specific numerical value can be set according to the actual situation;
Transmitting the generated steady state index to a comparison module;
The comparison module is used for comparing the steady state index with a preset steady state index threshold value and determining a repair plan of the prediction model;
if the steady state index is greater than or equal to the steady state index threshold, the hidden danger of the energy prediction model is continuously increased, obvious abnormal risks appear in the stability of the energy prediction model, an early warning signal is sent, and operations such as model repair and retraining are carried out after the prediction of the energy prediction model is finished, so that the stability and accuracy of the prediction are improved, and the potential energy saving opportunity is avoided being missed;
If the steady state index is smaller than the steady state index threshold, the hidden danger of the energy prediction model is possibly an accidental event, an early warning signal is not required to be sent, and the operation such as repairing can be carried out on the hidden danger regularly;
according to the method, the prediction process of the energy prediction model is monitored, the monitored data are preprocessed, a plurality of characteristic data related to the prediction process of the energy prediction model are extracted, a data processing model is built according to the plurality of characteristic data, a prediction hidden danger assessment index is generated, the prediction process of the energy prediction model is assessed, the prediction process with larger hidden danger is marked, a data analysis set is built according to the marked prediction hidden danger assessment index, the existing prediction hidden danger assessment index in the data analysis set is analyzed, a steady state index is generated, the steady state index is compared with a preset steady state index threshold value, a repair plan of the prediction model is determined, obvious abnormal risks appear in the stability of the energy prediction model, early warning signals are sent out, operations such as model repair and retraining are carried out after the prediction of the energy prediction model, the stability and accuracy of the prediction are improved, energy waste and decision errors caused by instability of the model are avoided, the prediction accuracy is improved, the energy saving potential is recognized, an effective energy saving strategy is formulated, and energy utilization efficiency is improved.
The above formulas are all formulas with dimensions removed and numerical values calculated, the formulas are formulas with a large amount of data collected for software simulation to obtain the latest real situation, and preset parameters in the formulas are set by those skilled in the art according to the actual situation.
The above embodiments may be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented in software, the above-described embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product comprises one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the processes or functions described in accordance with embodiments of the present application are produced in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from one website site, computer, server, or data center to another website site, computer, server, or data center by wired or wireless means (e.g., infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains one or more sets of available media. The usable medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium may be a solid state disk.
It should be understood that, in various embodiments of the present application, the sequence numbers of the foregoing processes do not mean the order of execution, and the order of execution of the processes should be determined by the functions and internal logic thereof, and should not constitute any limitation on the implementation process of the embodiments of the present application.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a read-only memory (ROM), a random access memory (random access memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The foregoing is merely illustrative of the present application, and the present application is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (8)
1. An integrated energy management system, characterized in that: the system comprises a prediction model monitoring module, a central processing module, an analysis module and a comparison module, wherein the modules are connected through signals;
the prediction model monitoring module is used for monitoring the prediction process of the energy prediction model, preprocessing the monitored data to obtain multiple characteristic data of the prediction process, and transmitting the multiple characteristic data to the central processing module;
The central processing module establishes a data processing model according to the multiple feature data, generates a predicted hidden danger assessment index, marks the current predicted hidden danger assessment index, and transmits the marked predicted hidden danger assessment index to the analysis module;
The analysis module is used for establishing a data analysis set according to the marked predicted hidden danger assessment indexes, analyzing the existing predicted hidden danger assessment indexes in the data analysis set, generating a steady-state index and transmitting the steady-state index to the comparison module;
And the comparison module is used for comparing the steady state index with a preset steady state index threshold value and determining a repair plan of the prediction model.
2. An integrated energy management system according to claim 1, wherein: the multiple feature data comprise prediction interval error data and prediction process data, the prediction interval error data comprise confidence interval deviation coefficients, the prediction process data comprise prediction load performance coefficients and cartoon abnormal time coefficients, and the confidence interval deviation coefficients, the prediction load performance coefficients and the cartoon abnormal time coefficients are respectively marked as ZX, FX and KD.
3. An integrated energy management system according to claim 2, wherein: the confidence interval deviation coefficient acquisition logic is as follows:
Calculating a confidence interval expressed by the following expression zx=yu±α×se, wherein ZX represents the confidence interval, i.e., zx= [ ZX 1,ZX2 ], YU represents the predicted value, α represents the confidence coefficient, and SE represents the prediction standard error, wherein Where MSE represents mean square error, x represents energy prediction model input value,/>Representing an average value of input values of the energy prediction model, wherein i is the number of times of prediction of the energy prediction model, and n is a positive integer;
comparing the actual observation value with the confidence interval of the predicted result, and calculating the confidence interval deviation coefficient when the actual observation value exceeds the confidence interval of the predicted result, wherein the expression is as follows Where PC represents the deviation value of the confidence interval of the actual observed value and the predicted result, and SJ represents the actual observed value.
4. An integrated energy management system according to claim 2, wherein: the acquisition logic of the predicted load performance coefficient is as follows:
The expression of the calculated load expression value is as follows Wherein YC represents the deviation value between the actual observed value and the predicted value, FU represents different load states, the predicted data quantity can be quantized, and beta represents a constant for adjusting or normalizing the calculation result;
comparing the load representation value with a preset load representation value threshold, marking the load representation value as an abnormal load representation value when the load representation value is larger than or equal to the load representation value threshold, and calculating a predicted load representation coefficient according to the abnormal load representation value Wherein j represents the number of predictions of the occurrence of abnormal load representation values, and m is a positive integer;
The capture logic of the katon anomaly time coefficient is as follows:
calculating the single predicted duty cycle of the break-in time, the calculation expression of which is as follows Where KD represents the katen time of the single prediction process and ZH represents the execution time of the single prediction;
Comparing the duty ratio of the click time with a preset duty ratio threshold of the click time, marking the duty ratio of the click time as an abnormal duty ratio of the click time when the duty ratio of the click time is larger than or equal to the duty ratio threshold of the click time, and calculating a click abnormal time coefficient according to the abnormal duty ratio of the click time, wherein the expression is as follows Wherein AK represents abnormal cartoon time duty ratio times, and ZC represents total times predicted by the energy prediction model.
5. An integrated energy management system according to claim 2, wherein: calculating according to the confidence interval deviation coefficient, the predicted load expression coefficient and the cartoon abnormal time coefficient to obtain a predicted hidden danger assessment index through a formula;
comparing the predicted hidden danger assessment index with a preset predicted hidden danger assessment index threshold;
If the predicted hidden danger assessment index is greater than or equal to the predicted hidden danger assessment index threshold, marking the generated predicted hidden danger assessment index as an abnormal predicted hidden danger assessment index, and transmitting the abnormal predicted hidden danger assessment index to an analysis module;
if the predicted hidden danger assessment index is smaller than the predicted hidden danger assessment index threshold, no other operation is needed.
6. An integrated energy management system according to claim 1, wherein: establishing a data analysis set according to the sequentially received abnormal prediction hidden danger assessment indexes, and marking the data analysis set as Wherein/>The method comprises the steps that sequentially received abnormal prediction hidden danger assessment indexes are represented, v represents sequence numbers of the sequentially received abnormal prediction hidden danger assessment indexes, and u is a positive integer;
Acquiring outlier information and interval time information of existing abnormal prediction hidden danger assessment indexes in a data analysis set, wherein the outlier information comprises abnormal prediction hidden danger assessment index outliers, and the interval time information comprises interval time differences; and respectively marking the outlier and the interval time of the abnormal prediction hidden danger assessment index as QA and NP.
7. The integrated energy management system of claim 6, wherein: the calculation expression of the outlier of the abnormality prediction hidden danger assessment index is as follows The method comprises the steps that an abnormal prediction hidden danger assessment index with a number u in a data analysis set is represented, PZ represents the average value of existing abnormal prediction hidden danger assessment indexes in the data analysis set, and ZB represents the standard deviation of the existing abnormal prediction hidden danger assessment indexes in the data analysis set;
The calculation expression of the interval time difference is as follows np= |t δ-tγ |, where t δ represents the reception time of the abnormality prediction hidden danger assessment index numbered u, and t γ represents the reception time of the abnormality prediction hidden danger assessment index numbered u-1.
8. The integrated energy management system of claim 7, wherein: calculating a steady state index according to the outlier of the abnormal prediction hidden danger evaluation index and the interval time difference through a formula;
Comparing the steady state index with a preset steady state index threshold;
if the steady state index is greater than or equal to the steady state index threshold, an early warning signal is sent out, and operations such as model repair and retraining are carried out after the prediction of the energy prediction model is finished;
if the steady state index is smaller than the steady state index threshold, the early warning signal is not required to be sent, and the operation such as repairing can be performed on the steady state index periodically.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410131713.5A CN118071168B (en) | 2024-01-31 | Comprehensive energy management system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410131713.5A CN118071168B (en) | 2024-01-31 | Comprehensive energy management system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118071168A true CN118071168A (en) | 2024-05-24 |
| CN118071168B CN118071168B (en) | 2024-10-01 |
Family
ID=
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101589262B1 (en) * | 2014-09-24 | 2016-01-27 | 현대건설 주식회사 | Building energy management method by realtime self-calibrating building energy model |
| CN116466168A (en) * | 2023-04-24 | 2023-07-21 | 江苏新博能源科技有限公司 | New energy management platform anomaly monitoring system and method based on cloud computing |
| CN116645010A (en) * | 2023-07-21 | 2023-08-25 | 金正大生态工程集团股份有限公司 | Chemical industry safety in production inspection system |
| CN116863997A (en) * | 2023-07-14 | 2023-10-10 | 天长市金羽禽业有限公司 | Animal model-based Sanhuang chicken breeding system and method |
| CN117192502A (en) * | 2023-11-06 | 2023-12-08 | 中华人民共和国连云港海事局后勤管理中心 | Anchor ship anchor-moving monitoring system based on target radar |
| CN117422405A (en) * | 2023-10-31 | 2024-01-19 | 新乡九和电子科技有限公司 | Electric power system information processing system based on electric power big data |
| CN117455242A (en) * | 2023-11-27 | 2024-01-26 | 南京华控创为信息技术有限公司 | Water conservancy management system based on digital twinning |
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101589262B1 (en) * | 2014-09-24 | 2016-01-27 | 현대건설 주식회사 | Building energy management method by realtime self-calibrating building energy model |
| CN116466168A (en) * | 2023-04-24 | 2023-07-21 | 江苏新博能源科技有限公司 | New energy management platform anomaly monitoring system and method based on cloud computing |
| CN116863997A (en) * | 2023-07-14 | 2023-10-10 | 天长市金羽禽业有限公司 | Animal model-based Sanhuang chicken breeding system and method |
| CN116645010A (en) * | 2023-07-21 | 2023-08-25 | 金正大生态工程集团股份有限公司 | Chemical industry safety in production inspection system |
| CN117422405A (en) * | 2023-10-31 | 2024-01-19 | 新乡九和电子科技有限公司 | Electric power system information processing system based on electric power big data |
| CN117192502A (en) * | 2023-11-06 | 2023-12-08 | 中华人民共和国连云港海事局后勤管理中心 | Anchor ship anchor-moving monitoring system based on target radar |
| CN117455242A (en) * | 2023-11-27 | 2024-01-26 | 南京华控创为信息技术有限公司 | Water conservancy management system based on digital twinning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109524139B (en) | Real-time equipment performance monitoring method based on equipment working condition change | |
| US11977375B2 (en) | Scalable systems and methods for assessing healthy condition scores in renewable asset management | |
| CN116345700B (en) | Energy consumption monitoring method and monitoring system for energy storage power station | |
| CN110738346A (en) | batch electric energy meter reliability prediction method based on Weibull distribution | |
| CN112632773B (en) | Reliability prediction method for wind turbine generator | |
| CN116388402B (en) | Abnormality alarm analysis method applied to power transformation equipment | |
| CN115796708B (en) | Big data intelligent quality inspection method, system and medium for engineering construction | |
| CN116739829B (en) | Big data-based power data analysis method, system and medium | |
| CN117808456A (en) | Equipment fault early warning method and device based on intelligent operation management | |
| CN112580858A (en) | Equipment parameter prediction analysis method and system | |
| CN116972910A (en) | Monitoring method and system for electrical equipment of thermal power plant | |
| CN116028887A (en) | Analysis method of continuous industrial production data | |
| CN117574180B (en) | Fuel production and emission system data correlation control management system | |
| CN117560300B (en) | Intelligent internet of things flow prediction and optimization system | |
| CN114442543A (en) | Computer monitoring method suitable for early warning of hydropower station fault | |
| CN115114124A (en) | Host risk assessment method and device | |
| CN116720983A (en) | Power supply equipment abnormality detection method and system based on big data analysis | |
| CN118071168B (en) | Comprehensive energy management system | |
| CN117009741A (en) | Abnormal analysis method and system based on generated energy | |
| CN116128690B (en) | Carbon emission cost value calculation method, device, equipment and medium | |
| CN116910677A (en) | Industrial instrument fault diagnosis method and system | |
| CN118071168A (en) | Comprehensive energy management system | |
| CN116910635A (en) | Pressure regulator fault diagnosis method and device based on federal learning and expansion convolution | |
| CN117977816B (en) | Intelligent power supply system for electric power safety | |
| CN118157325B (en) | Real-time monitoring method and system for new energy power |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |