CN115860351A - Method and system for determining energy-saving emission-reducing target of high-energy-consumption enterprise - Google Patents
Method and system for determining energy-saving emission-reducing target of high-energy-consumption enterprise Download PDFInfo
- Publication number
- CN115860351A CN115860351A CN202211153494.8A CN202211153494A CN115860351A CN 115860351 A CN115860351 A CN 115860351A CN 202211153494 A CN202211153494 A CN 202211153494A CN 115860351 A CN115860351 A CN 115860351A
- Authority
- CN
- China
- Prior art keywords
- enterprise
- energy
- energy consumption
- emission
- consumption data
- 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
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a method for determining energy-saving and emission-reducing targets of high-energy-consumption enterprises, which is characterized by comprising the following steps of: collecting multiple groups of energy consumption data of an enterprise; establishing an enterprise energy consumption data matrix Y; calculating the energy consumption level of the enterprise through a neural network; outputting the classification result of the historical energy consumption level of the enterprise, and training a neural network; determining the current energy consumption level of each enterprise in the current statistical period by using a neural network; and selecting and recommending an energy-saving and emission-reducing scheme in the template library for the high-energy-consumption enterprise according to the current energy consumption level. According to the invention, historical energy consumption data of an enterprise are fully utilized, energy consumption data of the enterprise are collected from multiple dimensions, a neural network algorithm is introduced, the historical energy consumption data of the enterprise are used for training and verification, a weight coefficient in the neural network is correspondingly adjusted according to a verification result, so that an operation result of the neural network algorithm conforms to the historical energy consumption grade condition of the enterprise, then real-time energy consumption data of the current enterprise are collected, and the energy consumption grade of the enterprise is calculated in real time by using the trained energy consumption data.
Description
Technical Field
The invention relates to an energy-saving emission-reducing target optimization method in an enterprise, in particular to a method and a system for determining an energy-saving emission-reducing target of a high-energy-consumption enterprise.
Background
With the development of national economic technology and the improvement of living standard of people, sustainable development is a common concern of countries all over the world at present, especially when extreme weather frequently occurs at present, and the goal of realizing carbon peak-to-peak carbon neutralization is the central importance of domestic energy management at present.
In the process of energy management, one of the difficult problems is how to determine a high-energy-consumption enterprise, because the power consumption of the enterprise fluctuates obviously along with time, capacity and socioeconomic operation order, and the energy consumption level of the enterprise is not in a stable state, it is one of the problems that needs to be solved urgently in current energy management to determine the high-energy-consumption enterprise.
For the prediction and statistics of the electricity consumption of regional enterprises, a large amount of research is already carried out, and economic models and theoretical methods comprise error correction models, neural network models and the like. However, the changes of the climate, the social consumption level, the supply chain capacity and the economic environment make the prediction mode of the traditional economic power prediction model difficult to accurately reflect the power trumpet level of the current enterprise. The main reason is that the prior art focuses too much on the power consumption of enterprises, but neglects the economic efficiency and carbon emission level of the power consumption of the enterprises, so that the analysis of the high-power-consumption enterprises in the prior art is single.
Disclosure of Invention
The method for determining the energy-saving and emission-reducing targets of the high-energy-consumption enterprises, provided by the invention, can be used for solving the problem of determining the energy-saving and emission-reducing targets of the high-energy-consumption enterprises in the prior art, so that the carbon emission whole period can be reasonably managed.
A method for determining energy conservation and emission reduction targets of high-energy-consumption enterprises is characterized by comprising the following steps:
step 1: collecting multiple groups of energy consumption data of an enterprise;
step 2: establishing an enterprise energy consumption data matrix Y;
and step 3: calculating the energy consumption level of the enterprise through a neural network;
and 4, step 4: outputting a classification result of the historical energy consumption level of the enterprise, and training a neural network;
and 5: determining the current energy consumption level of each enterprise in the current statistical period by utilizing a neural network;
step 6: and selecting and recommending an energy-saving and emission-reducing scheme in the template library for the high-energy-consumption enterprise according to the current energy consumption level.
Preferably, the method is further characterized by the plurality of sets of energy consumption data in step (1):
determining an accounting boundary of a high-energy-consumption enterprise, setting that the enterprise to be determined in an area A to be accounted has I families, and acquiring various energy consumption data of the enterprise I, wherein the various energy consumption data at least comprise:
(1) The monthly power consumption H of the enterprise in the power grid peak period of the past year is counted i ；
(2) The monthly power consumption W of the enterprise in the power grid power utilization trough period of the past year is counted i ；
(3) Calculating the monthly average electric energy consumption L of the enterprise in the past year i ；
(4) Statistics of monthly average carbon emission C of i enterprise in past year i ；
(5) Statistical industry label R of i enterprise i ；
(6) Statistics is carried out on the fact that carbon emission factor in raw materials purchased by i enterprises in the past year is higher than 1.5 tons of CO 2 Material fraction per ton of coal S i ；
(7) The enterprise dockingMonthly average logistics transportation carbon emission V of the past year i ；
(8) Carbon emission quota M purchased by the enterprise from the carbon emission trading market in the past year i
(9) The enterprise has an energy consumption level N determined by a local energy consumption management department in the past year i ；
Wherein I is more than or equal to 1 and less than or equal to I.
Preferably, the business is further characterized by a carbon emission allowance M purchased from a carbon emission trading market over the past year by the business i i The calculation method of (A) is as follows: the number V of the transport vehicles of the raw materials purchased by the enterprise i is And a transport distance V isd The fuel used is F i Carbon emission factor is E i And the method also comprises a transportation vehicle number V for transporting the finished products/semi-finished products to the market or the enterprises in the upstream and downstream industrial chains ir And a transport distance V ird The fuel used is D i Carbon emission factor of G i . At the same time, calculating the logistics traffic V i ＝∑V is *V isd *F i *E i +∑V ir *V ird *D i *G i 。
Preferably, the industry label assigns the first value P to an enterprise in a high energy consuming industry 1 And a second value P is assigned to enterprises in the low energy consumption industry 2 And a third value P for enterprises of other industries 3 。
Preferably, the enterprise energy consumption data matrix Y in step 2 is
Preferably, step 3 further comprises, after the step of,
each row of the enterprise energy consumption data matrix Y is respectively and correspondingly input to a single input node of the neural network, the hidden layer is at least one layer, and a node function j of the hidden layer is a linear migration function:
Z j (x)＝ω j1 *x 1 +ω j2 *x 2 +ω j3 *x 3 +ω j4 *x 4 +ω j5 *x 5 +ω j6 *x 6 +ω j7 *x 7 +ω j8 *x 8 +Δδ j 。
in the above formula, ω j1 、ω j2 、ω j3 、ω j4 、ω j5 、ω j6 、ω j7 、ω j8 Is the weight coefficient of the jth node of the hidden layer, wherein j is more than or equal to 1 and less than or equal to N, and N is the number of nodes of the hidden layer; delta delta j Node bias value, x, for the jth node of the hidden layer 1 ……x 8 H corresponding to i enterprise i 、W i 、L i 、C i 、R i 、S i 、V i 、M i I.e. each row of the enterprise energy consumption data matrix is input into Y of the neural network algorithm 1 To Y 8 . The initial values of the weight coefficient and the offset value may be assigned to corresponding values by the energy management section in each region.
Preferably, further, after hiding the layer, a migration function F is set, which is in the functional form:
wherein the input value of the transfer function F
Preferably, the energy saving and emission reduction scheme in the template library in the step 6 includes not only specific measures of the energy saving and emission reduction scheme, but also various types of energy consumption data corresponding to the template enterprise providing the energy saving and emission reduction scheme, and when the energy saving and emission reduction scheme is provided for the enterprise to be subjected to emission reduction, similarity calculation needs to be performed on the various types of energy consumption data of the template enterprise and the various types of energy consumption data of the enterprise to be subjected to emission reduction, and the optimal energy saving and emission reduction scheme is determined according to the similarity calculation result.
Preferably, the similarity calculation at least needs to calculate the monthly average power consumption in the peak period of the power grid power consumption, the monthly average power consumption in the valley period of the power grid power consumption, the monthly average power consumption and the monthly average carbon emissionAnd purchasing raw materials with carbon emission factor higher than 1.5 tons of CO 2 And comparing the values of six dimensions, namely the material occupation ratio of each ton of coal and the monthly logistics transportation carbon emission, and determining whether the energy saving and emission reduction scheme of the template enterprise is suitable for the enterprise to be subjected to emission reduction according to the comparison result.
The invention has the beneficial effects that: compared with the prior art that a large amount of time is spent on collecting energy consumption data lagging behind enterprises by manually visiting each enterprise in an area to determine the energy consumption data of the enterprises, historical energy consumption data of the enterprises are collected from multiple dimensions by fully utilizing the historical energy consumption data of the enterprises, a neural network algorithm is introduced, the historical energy consumption data of the enterprises are used for training and verifying, and the weight coefficient in the neural network is correspondingly adjusted according to a verification result, so that the operation result of the neural network algorithm conforms to the historical energy consumption grade condition of the enterprises, then the current real-time energy consumption data of the enterprises are collected, and the energy consumption grade of the enterprises is calculated in real time by utilizing the trained energy consumption data. Meanwhile, a proper energy-saving emission-reducing scheme can be pushed to the enterprise from the server side in real time according to the current energy consumption level of the enterprise, especially, the similarity between the energy consumption data of the enterprise and the energy consumption data of the template enterprise is considered, the energy-saving emission-reducing optimization direction is determined, the target of energy saving and emission reduction is realized, the carbon emission compliance and carbon quota transaction cost of the enterprise can be reduced, and the development of green circular economy is promoted.
Drawings
FIG. 1 is a schematic flow chart illustrating steps of a method for determining an energy-saving and emission-reducing target of a high-energy-consumption enterprise according to the present invention.
FIG. 2 is a schematic diagram of an embodiment of a neural network architecture for determining an energy consumption level of an enterprise in the present invention.
FIG. 3 is a schematic diagram of a system architecture of the method for determining the energy saving and emission reduction target of the high energy consuming enterprise according to the present invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings, the steps of which are shown in fig. 1.
Step 1: the method comprises the following steps of collecting multiple groups of energy consumption data of high-energy-consumption enterprises, specifically:
firstly, determining an accounting boundary of a high-energy-consumption enterprise, setting that the enterprise to be determined with the energy consumption level in an area A to be accounted has I families, and acquiring various energy consumption data of the I families, wherein the various energy consumption data at least comprise:
(1) The monthly power consumption H of the enterprise in the power grid peak period of the past year is counted i ；
Such as demarcating 8:00-22: and 00 is the local daily power consumption peak period, and further, various electric energy counting devices such as an enterprise intelligent electric meter or an electric energy sensor are utilized to count the daily average power consumption of the enterprise in the power consumption peak period, for example, taking the past year as a counting period, and calculating the monthly average power consumption of each enterprise in the power consumption peak period and then averaging the monthly average power consumption, namely the monthly average power consumption of the enterprise in the power consumption peak period, wherein I is more than or equal to 1 and less than or equal to I.
(2) And (4) counting the monthly average power consumption W of the i enterprise in the power grid power utilization trough period of the past year i ；
Such as demarcating a daily 22: 00-day 8:00 is the local daily power consumption trough period, for example, taking the past year as a statistical period, and calculating an average value after calculating the monthly power consumption of each enterprise in the power consumption trough period, namely the monthly average power consumption of the enterprise in the power consumption trough period.
(3) Calculating the monthly average electric energy consumption L of the enterprise in the past year i ；
(4) Statistics of monthly average carbon emission C of i enterprise in past year i ；
(5) Statistical industry label R of i enterprise i ；
The industry label can classify enterprises in industries such as electric power, steel, nonferrous metal, building material, petrochemical industry, chemical industry, building and the like as high-energy-consumption industries according to the definition of the country on the high-energy-consumption industries, and assigns a first numerical value P to the enterprises belonging to the industries 1 And the method relates to the adoption of wind energy, light energy, ocean energy, biomass energy, nuclear fusion energy and the likeEnterprises in industry are classified as low-energy-consumption industry P 2 And assigning a second numerical value to the enterprises belonging to the industry, and setting the enterprises of other industries to be undetermined to uniformly assign a third numerical value P to the enterprises belonging to the industry 3 ；
(6) Statistics is carried out on the fact that carbon emission factor in raw materials purchased by i enterprises in the past year is higher than 1.5 tons of CO 2 Specific value S of coal/ton material in all raw materials i 。
The method comprises the steps that the ratio can be input by an enterprise, or enterprise purchasing information can be directly collected after the enterprise is authorized, and different purchased raw materials are calculated according to a carbon emission factor conversion table established by the country;
(7) The carbon emission V of monthly logistics transportation in the past year of the enterprise docking i ；
The logistics transportation volume comprises but is not limited to the number V of the transportation vehicles of the raw materials purchased by the enterprise is And a transport distance V isd The fuel used is F i Carbon emission factor is E i And the method also comprises a transportation vehicle number V for transporting the finished products/semi-finished products to the market or enterprises in upstream and downstream industrial chains ir And a transport distance V ird The fuel used is D i Carbon emission factor of G i . At the same time, calculating the logistics traffic V i ＝∑V is *V isd *F i *E i +∑V ir *V ird *D i *G i ；
(8) Carbon emission quota M purchased by the enterprise from the carbon emission trading market over the past year i
(9) The enterprise has been rated by the local energy consumption management department for an energy consumption level N in the past year i 。
Preferably, the above-mentioned energy consumption levels can be divided into three levels: high energy consumption, medium energy consumption and low energy consumption levels, and assigning values to the three energy consumption levels, such as 3, 2 and 1 to the high energy consumption, medium energy consumption and low energy consumption levels respectively.
And 2, step: establishing an enterprise energy consumption data matrix:
after the energy consumption data are counted, the data are listed as an enterprise energy consumption data matrix Y
It should be noted that the enterprise energy consumption data matrix Y does not include the energy consumption level N i Because the set of data is suitable for subsequent training and adjustment of the node functions of the neural network to ensure that the classification results for the neural network are appropriate.
And 3, step 3: calculating the energy consumption level of the enterprise through a neural network;
as shown in fig. 2: inputting the enterprise energy consumption data matrix Y established in step 2 into a neural network matrix, as shown in fig. 1, where each row of the enterprise energy consumption data matrix corresponds to a single input node of the neural network, the hidden layer is at least one layer, and the node function j of the hidden layer is a linear offset function:
Z j (x)＝ω j1 *x 1 +ω j2 *x 2 +ω j3 *x 3 +ω j4 *x 4 +ω j5 *x 5 +ω j6 *x 6 +ω j7 *x 7 +ω j8 *x 8 +Δδ j 。
in the above formula, ω j1 、ω j2 、ω j3 、ω j4 、ω j5 、ω j6 、ω j7 、ω j8 Is the weight coefficient of the jth node of the hidden layer, wherein j is more than or equal to 1 and less than or equal to N, and N is the number of nodes of the hidden layer; delta delta j Node bias value, x, for the jth node of the hidden layer 1 ……x 8 H corresponding to enterprise i i 、W i 、L i 、C i 、R i 、S i 、V i 、M i I.e. each row value of the enterprise energy consumption data matrix is respectively input into Y of the neural network algorithm 1 To Y 8 . The initial values of the weight coefficient and the offset value may be assigned to corresponding values by the energy management section in each region, and preferably, the weight coefficient may be determined according to a local important consideration for energy management, such as the region weightPoint survey for purchasing carbon emission quota in carbon emission trading market can be used for dividing omega into j8 The value is set to be higher, and for example, the monthly power consumption of the power consumption peak period is mainly considered in the region, omega can be set j1 The setting is higher and the functions of the remaining hidden layer nodes are of the same form.
Meanwhile, after hiding the layer, a migration function F is set, which has the functional form:
wherein the input value of the transfer function F
And 4, step 4: outputting the classification result of the historical energy consumption level of the enterprise and according to the energy consumption level N of the enterprise in the previous year i And training each node of the neural network to ensure the accuracy of neural network classification.
Step 5, inputting various energy consumption data of the enterprise in the area in the current statistical period into a neural network, and determining the current energy consumption level of each enterprise in the current statistical period, wherein the enterprise with high energy consumption level is the high energy consumption enterprise to be optimized and reduced in emission;
and 6: selecting and recommending an energy-saving and emission-reducing scheme in a template library for a high-energy-consumption enterprise according to the current energy consumption level;
the energy-saving emission-reducing scheme in the template library not only comprises specific measures of the energy-saving emission-reducing scheme, but also comprises various energy consumption data corresponding to the template enterprises providing the energy-saving emission-reducing scheme. Therefore, when an energy saving and emission reduction scheme is provided for an enterprise, similarity calculation needs to be performed on various types of historical energy consumption data of the template enterprise and various types of energy consumption data of a high-energy-consumption enterprise to be optimized and subjected to emission reduction.
The similarity calculation at least needs to compare the monthly average power consumption of the power grid in the peak period of power consumption, the monthly average power consumption of the power grid in the valley period of power consumption, the monthly average carbon emission and the carbon emission factor in the purchased raw materials which is higher than 1.5 tons of CO 2 Material fraction per ton of coalAnd comparing the values of the six dimensions of the carbon emission amount of the monthly logistics transportation, and determining whether the energy-saving emission-reducing scheme of the template enterprise is suitable for the high-energy-consumption enterprise to be optimized for emission reduction according to the comparison result.
Preferably, the high energy consumption enterprise to be optimized can obtain the average monthly power consumption at the peak time of power grid power consumption, the average monthly power consumption at the valley time of power grid power consumption, the average monthly carbon emission and the carbon emission factor in raw material purchase of more than 1.5 tons of CO from at least comparison 2 At least one or more than one selected material ratio per ton of coal is compared with corresponding energy consumption data between the energy saving and emission reduction schemes adopted by the template enterprises, for example, the numerical values of the two parties are directly divided, and if the result is higher than a certain threshold value, the energy saving and emission reduction optimization scheme provided by the template enterprises can be adopted.
It should be noted that, the energy consumption level corresponding to the larger value in the result output by the migration function is selected as the energy consumption level of the enterprise. As shown in FIG. 2, the output result of the transfer function F is classified into Q 1 ，Q 2 Or Q 3 For example, if the value is greater than 0.7 according to the calculation result of the migration function F, the value is classified as Q 1 I.e., a high energy consuming enterprise, if the value is less than or equal to 0.7 and greater than 0.4, it is classified as Q 2 Namely, the enterprise is a medium energy consumption enterprise, and if the energy consumption is less than 0.4, the enterprise is considered to be a low energy consumption enterprise.
Several specific examples are provided below:
example 1:
the method comprises the steps of collecting groups of energy consumption data of a plurality of years past in an area A, for example, groups of energy consumption data of the past 10 years, using data of 7 years as training data, using data of the last 3 years as verification data, verifying nodes of the neural network, and ensuring that the neural network can be accurately classified according to energy consumption performance of enterprises. Of course, if the amount of historical data is insufficient, the time span of training samples can be reduced to meet the training requirement.
Example 2:
statistics may be applied to account for poor integrity of historical data in certain regionsAbove H i 、W i 、L i 、C i 、R i 、S i 、V i 、M i The statistical period of the energy consumption data is changed to half a year or a quarter, so that a sufficient amount of training data can be generated for training the neural network algorithm.
Example 3:
for each group of data in step 1 of the above technical solution, the data may be collected and uploaded by the enterprise itself, or a regional energy management department may set a special energy consumption information collecting device in each enterprise, as shown in fig. 3, the energy consumption information collecting device is connected to the electric meters of each enterprise in the region to collect monthly average power consumption H in peak power consumption period i And the monthly power consumption W of the power consumption in the trough period i Monthly average electric energy consumption L i (ii) a Meanwhile, the energy consumption information acquisition equipment is connected to the carbon emission monitoring equipment of the enterprise to acquire the monthly carbon emission C of the enterprise i (ii) a Collecting a carbon emission quota M purchased from a carbon emission trading market of a business through financial information of the business i And the purchased raw material information, and counting the carbon emission factor of the raw materials purchased by the enterprise in the past to be higher than 1.5 tons of CO 2 Material fraction per ton of coal S i 。
After the information is collected, the data are uploaded to a server side of an energy consumption management system, and an industry label R of a corresponding enterprise is stored in the server side i And the energy consumption level N of the enterprise as determined by the local energy consumption management department in the past year i . And then, training energy consumption grade classification according to all the data, and adjusting the weight value of the hidden layer node function according to the training condition so as to ensure that the classification result of the neural network algorithm is correct.
The energy consumption level condition of the enterprise can be updated and adjusted in real time according to various energy consumption data of the enterprise through a neural network algorithm, and the regional energy consumption management and the carbon emission condition of the enterprise after energy saving and emission reduction measures are adopted can be dealt with in time.
The present invention and its embodiments have been described above, and the description is not intended to be limiting, and the drawings are only one embodiment of the present invention, and the actual structure is not limited thereto. In summary, those skilled in the art should appreciate that they can readily use the disclosed conception and specific embodiments as a basis for designing or modifying other structures for carrying out the same purposes of the present invention without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (10)
1. A method for determining energy-saving and emission-reduction targets of high-energy-consumption enterprises is characterized by comprising the following steps:
step 1: collecting multiple groups of energy consumption data of enterprises;
step 2: establishing an enterprise energy consumption data matrix Y;
and 3, step 3: calculating the energy consumption level of the enterprise through a neural network;
and 4, step 4: outputting the classification result of the historical energy consumption level of the enterprise, and training a neural network;
and 5: determining the current energy consumption level of each enterprise in the current statistical period by using a neural network;
step 6: and selecting and recommending an energy-saving and emission-reducing scheme in the template library for the high-energy-consumption enterprise according to the current energy consumption level.
2. The method of claim 1, further characterized by the plurality of sets of energy consumption data in step (1):
determining an accounting boundary of a high-energy-consumption enterprise, setting that an enterprise to be determined in an area A to be accounted has I enterprises, and acquiring various energy consumption data of the I enterprises, wherein the various energy consumption data at least comprise:
(1) Statistics of power of i enterprise in the past yearMonthly power consumption H of grid power utilization peak period i ；
(2) And (4) counting the monthly average power consumption W of the i enterprise in the power grid power utilization trough period of the past year i ；
(3) Calculating the monthly average electric energy consumption L of the enterprise in the past year i ；
(4) Statistics of monthly average carbon emission C of the past year of the enterprise i i ；
(5) Statistical industry label R of i enterprise i ；
(6) Statistics is carried out on the fact that the carbon emission factor in the raw materials purchased by the enterprise i in the past year is higher than 1.5 tons of CO 2 Specific value S of coal/ton material in all raw materials i ；
(7) The carbon emission V of monthly logistics transportation in the past year of the enterprise docking i ；
(8) Carbon emission quota M purchased by the enterprise from the carbon emission trading market over the past year i
(9) The enterprise has an energy consumption level N determined by a local energy consumption management department in the past year i ；
Wherein I is more than or equal to 1 and less than or equal to I.
3. The method of claim 2, further characterized in that the carbon emission quota M purchased by the i-business from the carbon emission trading market over the past year i The calculation method is as follows: the number V of the transport vehicles of the raw materials purchased by the enterprise i is And a transport distance V isd The fuel used is F i Carbon emission factor is E i And the method also comprises a transportation vehicle number V for transporting the finished products and/or the semi-finished products to the market or the enterprises in the upstream and downstream industrial chains ir And a transport distance V ird The fuel used is D i Carbon emission factor is G i While calculating the logistics transportation volume V i ＝∑V is *V isd *F i *E i +∑V ir *V ird *D i *G i 。
4. The method of claim 2, further characterized in that the industry tag is to enterprise of an energy intensive industryAssigning a first value P 1 And a second value P is assigned to enterprises in the low energy consumption industry 2 And a third value P for enterprises of other industries 3 。
5. The method of claim 2, further characterized in that the enterprise energy consumption data matrix Y in step 2 is
6. The method of claim 3, further characterized in that step 3 further comprises,
each row of the enterprise energy consumption data matrix input nodes Y corresponds to a single input node which is input into the neural network, the hidden layer is at least one layer, and the node function j of the hidden layer is a linear offset function:
Z j (x)＝ω j1 *x 1 +ω j2 *x 2 +ω j3 *x 3 +ω j4 *x 4 +ω j5 *x 5 +ω j6 *x 6 +ω j7 *x 7 +ω j8 *x 8 +Δδ j
in the above formula, ω j1 、ω j2 、ω j3 、ω j4 、ω j5 、ω j6 、ω j7 、ω j8 Is the weight coefficient of the jth node of the hidden layer, wherein j is more than or equal to 1 and less than or equal to N, and N is the number of nodes of the hidden layer; delta delta j Node bias value, x, for the jth node of the hidden layer 1 ......x 8 H corresponding to i enterprise i 、W i 、L i 、C i 、R i 、S i 、V i 、M i The numerical value of each line of the enterprise energy consumption data matrix is respectively input into an input node Y of the neural network algorithm 1 To Y 8 The initial values of the weight coefficient and the offset value may be assigned to corresponding values by the energy management section in each region.
7. The method of claim 6, further characterized by, after the hidden layer, setting a migration function F of the functional form:
wherein the input value of the transfer function F
8. The method as claimed in claim 1, wherein the energy saving and emission reduction scheme in the template library in step 6 includes not only specific measures of the energy saving and emission reduction scheme, but also various types of energy consumption data corresponding to a template enterprise providing the energy saving and emission reduction scheme, and when the energy saving and emission reduction scheme is provided for an enterprise to be emission reduced, similarity calculation needs to be performed on the various types of energy consumption data of the template enterprise and the various types of energy consumption data of the enterprise to be emission reduced, and an optimal energy saving and emission reduction scheme is determined according to a similarity calculation result.
9. The method of claim 8, further characterized in that the similarity calculation requires at least a monthly average power consumption during peak periods of grid power usage, a monthly average power consumption during valley periods of grid power usage, a monthly average power consumption, a monthly average carbon emission, a carbon emission factor in procurement of raw materials of greater than 1.5 tons CO 2 And carrying out numerical value comparison on six dimensions of the material occupation ratio of each ton of coal and the monthly logistics transportation carbon emission, and determining whether the energy-saving emission-reducing scheme of the template enterprise is suitable for the enterprise to be subjected to emission reduction according to the comparison result.
10. A system applying the method for determining energy conservation and emission reduction targets of energy-consuming enterprises as claimed in claim 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211153494.8A CN115860351A (en) | 2022-09-21 | 2022-09-21 | Method and system for determining energy-saving emission-reducing target of high-energy-consumption enterprise |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211153494.8A CN115860351A (en) | 2022-09-21 | 2022-09-21 | Method and system for determining energy-saving emission-reducing target of high-energy-consumption enterprise |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115860351A true CN115860351A (en) | 2023-03-28 |
Family
ID=85661069
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211153494.8A Pending CN115860351A (en) | 2022-09-21 | 2022-09-21 | Method and system for determining energy-saving emission-reducing target of high-energy-consumption enterprise |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115860351A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116485210A (en) * | 2023-06-14 | 2023-07-25 | 红杉天枰科技集团有限公司 | Neural network-based method and device for generating emission reduction strategy of agricultural management activity |
| CN117217966A (en) * | 2023-09-26 | 2023-12-12 | 广州市城市规划勘测设计研究院 | Intelligent electricity management system and method based on carbon emission and energy consumption |
| CN117808216A (en) * | 2024-03-01 | 2024-04-02 | 四川省铁路建设有限公司 | Energy saving and emission reduction effect evaluation method for sewage treatment |
-
2022
- 2022-09-21 CN CN202211153494.8A patent/CN115860351A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116485210A (en) * | 2023-06-14 | 2023-07-25 | 红杉天枰科技集团有限公司 | Neural network-based method and device for generating emission reduction strategy of agricultural management activity |
| CN116485210B (en) * | 2023-06-14 | 2023-09-05 | 红杉天枰科技集团有限公司 | Neural network-based method and device for generating emission reduction strategy of agricultural management activity |
| CN117217966A (en) * | 2023-09-26 | 2023-12-12 | 广州市城市规划勘测设计研究院 | Intelligent electricity management system and method based on carbon emission and energy consumption |
| CN117808216A (en) * | 2024-03-01 | 2024-04-02 | 四川省铁路建设有限公司 | Energy saving and emission reduction effect evaluation method for sewage treatment |
| CN117808216B (en) * | 2024-03-01 | 2024-05-07 | 四川省铁路建设有限公司 | Energy saving and emission reduction effect evaluation method for sewage treatment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115860351A (en) | Method and system for determining energy-saving emission-reducing target of high-energy-consumption enterprise | |
| Madhukumar et al. | Regression model-based short-term load forecasting for university campus load | |
| CN105160416A (en) | Transformer area reasonable line loss prediction method based on principal component analysis and neural network | |
| CN112734101B (en) | Sharing bicycle intelligent allocation method based on vehicle demand prediction | |
| WO2021135727A1 (en) | Energy data warehouse system | |
| CN102663065A (en) | Method for identifying and screening abnormal data of advertising positions | |
| Liu et al. | China's Export Surge and the New Margins of Trade | |
| Livas-García et al. | Forecasting of locational marginal price components with artificial intelligence and sensitivity analysis: A study under tropical weather and renewable power for the mexican southeast | |
| CN110826827B (en) | Enterprise online energy auditing system and method based on energy internet | |
| CN116433440A (en) | Data autoregressive enhanced carbon emission measuring and calculating method, system and electronic equipment | |
| CN114676931B (en) | Electric quantity prediction system based on data center technology | |
| CN115049314A (en) | Method for optimizing scale of self-contained fleet of main line transportation | |
| Fathabad et al. | Data-Driven Optimization for E-Scooter System Design | |
| Wu et al. | Risk Assessment of Low‐Speed Wind Power Projects Based on an Aggregated Cloud Method: A Case in China | |
| Zhao et al. | Forecasting regional short-term freight volume using QPSO-LSTM algorithm from the perspective of the importance of spatial information | |
| CN112001551A (en) | Method for predicting electricity sales amount of power grid in city based on electricity information of large users | |
| CN115375159B (en) | Real-time accounting method and system for total urban carbon emission | |
| CN117236532B (en) | Load data-based electricity consumption peak load prediction method and system | |
| Töppel | Risk and Return Management for the Heat Transition in Germany | |
| Dong et al. | Research on efficiency measurement of information industry chain integration based on multiple structures and its application in carbon management industrial park | |
| Yang et al. | Sustainable supply chain network design considering the interactive influence of the multiproduct on the production capacity | |
| Wang et al. | Research on Daily Tourist Flow Prediction of Scenic Spots Based on Similar Day Clustering and LSSVM Model | |
| Pawar et al. | Electricity Forecasting Using Machine Learning: A Review | |
| Saber | Quantifying forecast uncertainty in the energy domain | |
| Zhang et al. | An evaluation method for charging facilities siting |
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 |