CN116402187A - Enterprise pollution discharge prediction method based on power big data - Google Patents
Enterprise pollution discharge prediction method based on power big data Download PDFInfo
- Publication number
- CN116402187A CN116402187A CN202310145565.8A CN202310145565A CN116402187A CN 116402187 A CN116402187 A CN 116402187A CN 202310145565 A CN202310145565 A CN 202310145565A CN 116402187 A CN116402187 A CN 116402187A
- Authority
- CN
- China
- Prior art keywords
- data
- pollution discharge
- historical
- enterprise
- power
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 25
- 230000005611 electricity Effects 0.000 claims abstract description 48
- 238000007477 logistic regression Methods 0.000 claims abstract description 27
- 238000004140 cleaning Methods 0.000 claims abstract description 14
- 230000001419 dependent effect Effects 0.000 claims abstract description 8
- 238000007781 pre-processing Methods 0.000 claims abstract description 4
- 239000010865 sewage Substances 0.000 claims description 29
- 238000012549 training Methods 0.000 claims description 18
- 238000012544 monitoring process Methods 0.000 claims description 16
- 238000001514 detection method Methods 0.000 claims description 14
- 230000009471 action Effects 0.000 claims description 11
- 230000008569 process Effects 0.000 claims description 5
- 230000002159 abnormal effect Effects 0.000 claims description 4
- 238000012360 testing method Methods 0.000 claims description 3
- 238000003911 water pollution Methods 0.000 abstract description 3
- 208000028659 discharge Diseases 0.000 description 47
- 238000007726 management method Methods 0.000 description 7
- 238000013145 classification model Methods 0.000 description 3
- 238000009826 distribution Methods 0.000 description 3
- 230000007613 environmental effect Effects 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000007635 classification algorithm Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000002354 daily effect Effects 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000004880 explosion Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000013058 risk prediction model Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Images
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
- G06Q10/04—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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/27—Regression, e.g. linear or logistic regression
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Abstract
The invention discloses an enterprise pollution discharge prediction method based on power big data, which comprises the following steps: basic information acquisition: acquiring historical electricity consumption data and historical pollution discharge data of an enterprise; data preprocessing: carrying out data cleaning on the historical electricity consumption data according to the historical pollution discharge data; and (3) constructing characteristic indexes: extracting multidimensional electricity utilization characteristic indexes from the historical electricity utilization data after data cleaning; constructing a user pollution discharge prediction model: taking the electricity utilization characteristic index as an independent variable, taking the historical pollution discharge data as an independent variable, taking the independent variable as an input, taking the dependent variable as an output to train a logistic regression model, and taking the trained logistic regression model as an enterprise pollution discharge prediction model. The invention predicts the pollution discharge condition of potential water pollution enterprises and saves the manpower and the time of on-site checking.
Description
Technical Field
The invention relates to the technical field of electric power, in particular to an enterprise pollution discharge prediction method based on electric power big data.
Background
At present, the overall quality of the national water environment is kept continuously improved, key pollution discharge units in the convection area are required to be routinely detected in surrounding cities of the water area, and the consumption of manpower and material resources is high.
With smart grids and informatization construction, the power industry has accumulated massive data, which have characteristics of big data in terms of data volume, diversity, speed and value. The power industry has entered the big data age. The large power data is a mass structured, semi-structured and unstructured service data set collected through various data acquisition channels such as sensors, intelligent equipment, video monitoring equipment, audio communication equipment, mobile terminals and the like. How to analyze and early warn the pollution discharge situation of enterprises by means of the large electric power data is a considerable problem.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: aiming at the technical problems in the prior art, the invention provides the enterprise pollution discharge prediction method based on the power big data, which can predict the pollution discharge condition of a potential water pollution enterprise, help to improve the supervision efficiency and save the manual on-site check times.
In order to solve the technical problems, the technical scheme provided by the invention is as follows:
an enterprise pollution discharge prediction method based on power big data comprises the following steps:
step 3, constructing characteristic indexes: extracting multidimensional electricity utilization characteristic indexes from the historical electricity utilization data after data cleaning;
step 4, constructing a user pollution discharge prediction model: taking the electricity utilization characteristic index as an independent variable, taking the historical pollution discharge data as an independent variable, taking the independent variable as an input, taking the dependent variable as an output to train a logistic regression model, and taking the trained logistic regression model as an enterprise pollution discharge prediction model.
Further, the historical electricity consumption data comprises a user name, a user number, a contract capacity, a metering point number, an acquisition time point, a sharp electric quantity, a peak electric quantity, gu Dianliang, a flat electric quantity, a time point load, a time point current, a time point voltage and a time point power; the historical pollution discharge data comprises enterprise names, sewage detection time and detection results.
Further, step 2 specifically includes: according to the sewage detection time and the detection result in the historical sewage data, determining a time period with a sewage monitoring result, extracting power utilization data corresponding to the time period with the sewage monitoring result in the historical power utilization data, deleting missing data in the power utilization data, and adopting a box line graph to reject abnormal values in the power utilization data.
Further, the step 3 specifically includes: and determining a business target and formulating an action strategy by using an OSM model, and constructing and evaluating an index of historical electricity utilization data after data cleaning according to the action strategy to obtain a multidimensional electricity utilization characteristic index.
Further, the specific steps of constructing and evaluating the index of the historical electricity consumption data after the data cleaning according to the action strategy include:
drawing trend graphs of the standard exceeding sample enterprises and the non-standard exceeding sample enterprises in the same period with respect to the data for the first type of data, observing the trend graphs, and splitting the first type of data into corresponding indexes according to the observation results;
for the second type of data, the data is directly used as one type of index according to expert experience.
Further, the data amount of the first type of data is larger than a preset threshold value.
Further, in step 4, the electricity utilization characteristic index of the independent variable is staggered with the time of the historical pollution discharge data of the corresponding independent variable, and the time of the electricity utilization characteristic index is before the time of the corresponding historical pollution discharge data.
Further, the expression of the logistic regression model in step 4 is as follows:
in the above formula, P (ω, b) represents the enterprise pollution discharge probability, x represents the feature vector of the input sample of the logistic regression model, ω represents the weight vector composed of the weight values of each input feature, and b represents the deviation of the regression model.
Further, training the logistic regression model in step 4 includes: in the model fitting process, the model is optimized by adjusting the classification weight, and after model fitting is completed, the model with larger recall rate is accepted under the condition of the same accuracy rate.
Further, training the logistic regression model in step 4 includes: both independent and dependent variables were calculated as 3:1, setting a classification mode in training parameters as a plurality of modes, and setting weights of various types in the training parameters as 0:0.25 and 1:0.75, wherein 0 represents that an enterprise pollution discharge result is not pollution discharge, 1 represents that the enterprise pollution discharge result is pollution discharge, and using a training set training logistic regression model.
Compared with the prior art, the invention has the advantages that:
the invention is different from the traditional sewage collection monitoring and installing additional hardware facilities, does not need to modify hardware equipment, innovatively uses electric power data to assist the enterprise in monitoring and pollution discharge prediction, also constructs brand-new electric characteristic indexes according to service requirements and electric power data, and finally constructs an enterprise pollution discharge risk prediction model by using a logistic regression algorithm, and calculates the pollution discharge possibility of the enterprise through the model.
Drawings
FIG. 1 is a flow chart of a method according to an embodiment of the invention.
Fig. 2 is a load factor trend chart of an over-standard sample enterprise and an under-standard sample in the same period in an embodiment of the present invention.
FIG. 3 is a schematic diagram of enterprise pollution risk prediction in an embodiment of the present invention.
FIG. 4 is a diagram of a model predictive result confusion matrix in an embodiment of the invention.
Detailed Description
The invention is further described below in connection with the drawings and the specific preferred embodiments, but the scope of protection of the invention is not limited thereby.
Before describing the specific scheme of the embodiment, the following description is given to related contents:
the 2021 national power grid company is greatly advanced to digital transformation, the power data value-added service is enhanced, the power data value is deeply mined by applying a big data technology, and a power big data and river basin environment-friendly monitoring mode is constructed. Based on analysis prediction capability of big data, the pollution discharge condition of a potential water pollution enterprise is analyzed and early-warned on line, a digital basis is provided, the supervision efficiency is improved in an auxiliary mode, and manual on-site checking times are saved. The big data of electric power mainly comes from various links such as power generation, power transmission, transformation, power distribution, electricity consumption, scheduling and the like, and can be roughly divided into three types:
grid operation, equipment detection or monitoring data. The system mainly comprises an energy management system, a distribution network management system, a wide area measurement management system, a production management system, a power grid dispatching management system, a fault management system, an image monitoring system and the like.
Marketing data of the power enterprises, such as data of transaction electricity price, electricity sales quantity, electricity utilization clients and the like. The system mainly comprises a marketing business system, a 95598 customer service system, an electric energy metering system, an electricity consumption information acquisition system and the like.
And (5) power enterprise management data. Mainly in collaborative office systems, enterprise resource planning systems (ERP), etc.
The embodiment provides an enterprise pollution discharge prediction method based on power big data, as shown in fig. 1, comprising the following steps:
step 3, constructing characteristic indexes: extracting multidimensional electricity utilization characteristic indexes from the historical electricity utilization data after data cleaning;
step 4, constructing a user pollution discharge prediction model: taking the electricity utilization characteristic index as an independent variable, taking the historical pollution discharge data as an independent variable, taking the independent variable as an input, taking the dependent variable as an output to train a logistic regression model, and taking the trained logistic regression model as an enterprise pollution discharge prediction model.
In step 1 of the embodiment, considering that the power consumption data of the enterprise is mainly based on the data tables of user information, business expansion, power consumption, load, current, voltage, power and the like, the historical power consumption data includes key fields of user name, user number, contract capacity, metering point number, collection time point, sharp power, peak power, gu Dianliang, flat power, time point load, time point current, time point voltage, time point power and the like; the sewage disposal data of enterprises is mainly based on a historical sewage disposal monitoring result table provided by the environmental protection agency, and the historical sewage disposal data comprises fields such as enterprise names, sewage detection time, detection results (whether exceeding standards or not) and the like. The method comprises the steps that raw data are obtained from a marketing service application system, an electricity consumption information acquisition system, an environmental protection agency pollution discharge monitoring system and the like, wherein the marketing service application system mainly extracts archives data such as users, metering points and the like; the electricity consumption information acquisition system mainly extracts electric quantity, load, current, voltage and other acquisition data; the environmental protection agency sewage monitoring system mainly extracts the historical sewage condition of enterprises.
Step 3 of the embodiment uses an OSM model to construct a multidimensional index based on power data such as enterprise power consumption, enterprise load data, enterprise voltage data, enterprise line loss data and the like, and differentially represents power consumption characteristics between users and users without pollution discharge. The OSM model comprises three aspects of business targets, action strategies and evaluation indexes, and specifically comprises the following steps:
and 3.1, establishing a business target, namely using the simplest index to represent the electricity utilization difference between pollution discharge users to the greatest extent.
And 3.2, formulating an action strategy, namely an index formulation principle by the business target, wherein the principle of combining historical data with business understanding is adopted in the embodiment. The principle of the historical data specifically refers to historical data analysis, and is applied to the dimension with more data quantity so as to avoid the situation that the dimension explosion occurs when the data quantity is more and the data quantity is directly used as an input index; the principle of service understanding specifically refers to service understanding indexes, specifically refers to that certain kind of data is directly used as one of input indexes according to past experience summary of service personnel.
And 3.3, determining a business target and formulating an action strategy by using an OSM model, and constructing and evaluating the index of the historical electricity utilization data after data cleaning according to the action strategy to obtain a multidimensional electricity utilization characteristic index. The specific process of constructing and evaluating the index of the historical electricity consumption data after data cleaning according to the action strategy is as follows:
based on a principle of historical data analysis, drawing trend graphs of an exceeding sample enterprise and a non-exceeding sample enterprise in the same period on the first type of data, observing the trend graphs, and splitting the first type of data into corresponding indexes according to observation results;
based on the principle of service understanding index, for the second class of data, the data is directly used as one class of index according to expert experience.
The first type of data takes load as an example, the data has 96 points collected every day, the data volume is large, and the condition of dimensional explosion can occur when the data is directly used as an input index, so that the data is converted into daily load rate as an index; however, when the period of pollution discharge monitoring is monthly or quarterly, the number of daily load rates is also large, so that a new index capable of profiling load characteristics in the period is constructed, and a load rate trend chart of an exceeding sample enterprise and a non-exceeding sample in the same period is firstly drawn, as shown in fig. 2. By observing fig. 2, it is found that the load rate fluctuation of the exceeding sample is smaller and is within the interval of 40% -80%, probably because the pollution discharge exceeding enterprise production has no obvious peaks and valleys, and the load rate of the non-exceeding sample enterprise is less than 20% for many days except for larger fluctuation. Based on such findings, there is a great difference in mean, maximum, minimum, and distribution of load trends of non-superscalar users and superscalar users, and thus it is necessary to split and convert loads into indexes more adapted to models.
The second class of data takes contract capacity as an example, and can be directly used as one of input indexes according to past experience summary of service personnel. In this embodiment, the first type of data and the second type of data may be distinguished by setting a data amount threshold, where the data amount of the first type of data is greater than a preset threshold.
By means of the step 3 of the process, the electricity consumption characteristic index in this embodiment is shown in the following table:
table 1 input index of pollution discharge prediction model for enterprises
Step 4 of the embodiment utilizes a logistic regression algorithm to construct an enterprise pollution discharge prediction model, and then adopts logistic regression to predict specifically comprises the following steps:
in step 4.1, the electricity consumption characteristic index obtained in step 3 is taken as an independent variable, and the historical pollution discharge data is taken as an independent variable, as shown in fig. 3, in this embodiment, different independent variables are divided according to time periods, the independent variables corresponding to the independent variables are also divided according to time periods, in order to predict future pollution discharge conditions of enterprises, the model does not adopt traditional synchronous input and output, but the time of the electricity consumption characteristic index of the independent variable is staggered with that of the historical pollution discharge data corresponding to the independent variables, and because the output historical pollution discharge data is divided in quarternary units, the input electricity consumption characteristic index adopts the electricity consumption characteristic of the month before the prediction target, so that the accuracy of model data is improved, namely, the time of the electricity consumption characteristic index is before the time of the corresponding historical pollution discharge data.
Step 4.2, in order to mine hidden connection between historical electricity utilization characteristics and future pollution discharge conditions and obtain pollution discharge risk degree, a logistic regression is drawn as a target model, and the expression of the logistic regression model is as follows:
in the above formula, P (ω, b) represents the enterprise pollution discharge probability, x represents the feature vector of the input sample of the logistic regression model, ω represents the weight vector composed of the weight values of each input feature, and b represents the deviation of the regression model.
Step 4.3, training a logistic regression model, as shown in fig. 3, specifically including:
and 4.3.1, optimizing the model by adjusting the classification weight in the model fitting process. Because it is often more preferable to accept models with higher recall rates for a given classification weight than for an enterprise with a blowdown that is identified as non-superscalar.
And 4.3.2, when the model is evaluated after fitting, receiving the model with larger recall rate under the condition of the same accuracy rate according to the service requirement.
Step 4.3.3, using the sample data to complete the instantiation: both independent and dependent variables were calculated as 3:1 and dividing the training set and the test set by using the training set to train the logistic regression model. The training parameters were finally set as follows:
table 2 network parameter search scope
| Parameter name | Parameter value |
| class_weight (various types of weights in classification model) | 0:0.25；1:0.75 |
| Multiclass (Classification mode) | multinomial |
The class_weight parameter is used for marking various weights in the classification model (namely the logistic regression model), the weights need to be adjusted according to input data, and the final values are 0:0.25 and 1:0.75, wherein 0 represents that the sewage disposal result of an enterprise is not sewage disposal, and 1 represents that the sewage disposal result of the enterprise is sewage disposal; the classification mode in the multi_class parameter classification model is a multi-mode.
And storing the model after training, wherein the result accuracy of the prediction test set reaches 90.32%, and the confusion matrix is shown in fig. 4.
In summary, the invention has the following beneficial technical effects:
(1) And the enterprise pollution discharge risk prediction analysis based on big data is different from the traditional pollution discharge detection based on hardware instruments, and is beneficial to saving equipment cost and labor cost.
(2) The OSM model is combined with the machine learning classification algorithm, optimization is carried out based on the OSM model in the index construction and data classification algorithm, and compared with the traditional model, the method can be closer to service requirements, and the prediction accuracy is improved.
The foregoing is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. While the invention has been described with reference to preferred embodiments, it is not intended to be limiting. Therefore, any simple modification, equivalent variation and modification of the above embodiments according to the technical substance of the present invention shall fall within the scope of the technical solution of the present invention.
Claims (10)
1. An enterprise pollution discharge prediction method based on electric power big data is characterized by comprising the following steps:
step 1, basic information acquisition: acquiring historical electricity consumption data and historical pollution discharge data of an enterprise;
step 2, data preprocessing: carrying out data cleaning on the historical electricity consumption data according to the historical pollution discharge data;
step 3, constructing characteristic indexes: extracting multidimensional electricity utilization characteristic indexes from the historical electricity utilization data after data cleaning;
step 4, constructing a user pollution discharge prediction model: taking the electricity utilization characteristic index as an independent variable, taking the historical pollution discharge data as an independent variable, taking the independent variable as an input, taking the dependent variable as an output to train a logistic regression model, and taking the trained logistic regression model as an enterprise pollution discharge prediction model.
2. The method for predicting pollution discharge of an enterprise based on big power data as claimed in claim 1, wherein the historical power consumption data includes user name, user number, contract capacity, metering point number, collection time point, sharp power, peak power, gu Dianliang, flat power, time point load, time point current, time point voltage, and time point power; the historical pollution discharge data comprises enterprise names, sewage detection time and detection results.
3. The method for predicting pollution discharge of an enterprise based on electric power big data as set forth in claim 1, wherein the step 2 specifically includes: according to the sewage detection time and the detection result in the historical sewage data, determining a time period with a sewage monitoring result, extracting power utilization data corresponding to the time period with the sewage monitoring result in the historical power utilization data, deleting missing data in the power utilization data, and adopting a box line graph to reject abnormal values in the power utilization data.
4. The method for predicting pollution discharge of an enterprise based on electric power big data as set forth in claim 1, wherein the step 3 specifically includes: and determining a business target and formulating an action strategy by using an OSM model, and constructing and evaluating an index of historical electricity utilization data after data cleaning according to the action strategy to obtain a multidimensional electricity utilization characteristic index.
5. The method for predicting pollution discharge of an enterprise based on big electric power data as set forth in claim 4, wherein the specific steps of constructing and evaluating the historical electric power consumption data after data cleaning according to the action strategy include:
drawing trend graphs of the standard exceeding sample enterprises and the non-standard exceeding sample enterprises in the same period with respect to the data for the first type of data, observing the trend graphs, and splitting the first type of data into corresponding indexes according to the observation results;
for the second type of data, the data is directly used as one type of index according to expert experience.
6. The method for predicting emissions of an enterprise based on electrical power big data of claim 5, wherein the data size of the first type of data is greater than a preset threshold.
7. The method of claim 1, wherein the electricity consumption characteristic index of the independent variable in step 4 is staggered in time from the historical blowdown data of the corresponding dependent variable, and the time of the electricity consumption characteristic index is before the time of the corresponding historical blowdown data.
8. The method for predicting pollution discharge of an enterprise based on electric power big data as set forth in claim 1, wherein the expression of the logistic regression model in step 4 is as follows:
in the above formula, P (ω, b) represents the enterprise pollution discharge probability, x represents the feature vector of the input sample of the logistic regression model, ω represents the weight vector composed of the weight values of each input feature, and b represents the deviation of the regression model.
9. The method of claim 1, wherein training the logistic regression model in step 4 comprises: in the model fitting process, the model is optimized by adjusting the classification weight, and after model fitting is completed, the model with larger recall rate is accepted under the condition of the same accuracy rate.
10. The method of claim 1, wherein training the logistic regression model in step 4 comprises: both independent and dependent variables were calculated as 3:1, dividing a training set and a test set in proportion, setting a classification mode in training parameters as a multi-term mode, and setting classification weights in the training parameters as 0:0.25;1:0.75, wherein 0 represents that the sewage disposal result of the enterprise is not sewage disposal, 1 represents that the sewage disposal result of the enterprise is sewage disposal, and a training set training logistic regression model is used.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310145565.8A CN116402187A (en) | 2023-02-21 | 2023-02-21 | Enterprise pollution discharge prediction method based on power big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310145565.8A CN116402187A (en) | 2023-02-21 | 2023-02-21 | Enterprise pollution discharge prediction method based on power big data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116402187A true CN116402187A (en) | 2023-07-07 |
Family
ID=87016751
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310145565.8A Pending CN116402187A (en) | 2023-02-21 | 2023-02-21 | Enterprise pollution discharge prediction method based on power big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116402187A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117609926A (en) * | 2024-01-23 | 2024-02-27 | 中科三清科技有限公司 | Pollution discharge mechanism production state determining method and device based on power data |
| CN117829614A (en) * | 2024-03-06 | 2024-04-05 | 四川国蓝中天环境科技集团有限公司 | Industrial enterprise pollution discharge risk classification calculation method based on multi-source data fusion |
-
2023
- 2023-02-21 CN CN202310145565.8A patent/CN116402187A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117609926A (en) * | 2024-01-23 | 2024-02-27 | 中科三清科技有限公司 | Pollution discharge mechanism production state determining method and device based on power data |
| CN117609926B (en) * | 2024-01-23 | 2024-04-16 | 中科三清科技有限公司 | Pollution discharge mechanism production state determining method and device based on power data |
| CN117829614A (en) * | 2024-03-06 | 2024-04-05 | 四川国蓝中天环境科技集团有限公司 | Industrial enterprise pollution discharge risk classification calculation method based on multi-source data fusion |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116402187A (en) | Enterprise pollution discharge prediction method based on power big data | |
| Li et al. | Predicting short-term electricity demand by combining the advantages of arma and xgboost in fog computing environment | |
| CN110097297A (en) | A kind of various dimensions stealing situation Intellisense method, system, equipment and medium | |
| CN116646933A (en) | Big data-based power load scheduling method and system | |
| CN105426980B (en) | Power distribution network health index assessment engineering application system | |
| CN106570581A (en) | Attribute association based load prediction system and method in energy Internet environment | |
| CN110222991B (en) | Metering device fault diagnosis method based on RF-GBDT | |
| CN105373971A (en) | Method of building energy efficiency management on the basis of big data | |
| CN113064866B (en) | Power business data integration system | |
| CN103020459A (en) | Method and system for sensing multiple-dimension electric utilization activities | |
| CN112688431A (en) | Power distribution network load overload visualization method and system based on big data | |
| CN116073372A (en) | Intelligent electricity utilization-based safety monitoring management system and method | |
| CN103617447A (en) | Evaluation system and method for intelligent substation | |
| CN115423429A (en) | Multimode integrated distribution network operation system based on image and sound information | |
| CN115358522A (en) | Enterprise online monitoring system and method | |
| CN112508254B (en) | Method for determining investment prediction data of transformer substation engineering project | |
| CN113902181A (en) | Short-term prediction method and equipment for common variable heavy overload | |
| CN111552686B (en) | Power data quality assessment method and device | |
| CN112633762A (en) | Building energy efficiency obtaining method and equipment | |
| CN115905319B (en) | Automatic identification method and system for abnormal electricity fees of massive users | |
| CN116993227A (en) | Heat supply analysis and evaluation method, system and storage medium based on artificial intelligence | |
| Huaiying et al. | Research on technical architecture and application of big data cloud platform for electric power measurement | |
| CN114662563A (en) | Industrial electricity non-invasive load decomposition method based on gradient lifting algorithm | |
| CN108123436B (en) | Voltage out-of-limit prediction model based on principal component analysis and multiple regression algorithm | |
| Keyan et al. | Anomaly detection method of distribution network line loss based on hybrid clustering and LSTM |
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 |