Carbon emission prediction method, device and electronic equipment
By constructing a generator set load prediction model and raw coal consumption rate and plant power prediction model, combining macro factors and generator set equipment data, predicting carbon emissions of power generation enterprises, the problem of low prediction accuracy in the existing technology is solved, and more accurate carbon emission prediction and production optimization guidance is achieved.
Patent Information
- Application Number
- CN202311664936.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-12-05
- Publication Date
- 2025-05-09
- Estimated Expiration
- 2043-12-05
AI Technical Summary
The existing technology only considers macro factors to predict the total carbon emissions in a region, and fails to predict carbon emissions based on the actual situation of the equipment of the power generation enterprise, resulting in low prediction accuracy and cannot provide guiding significance for the production optimization of the power generation enterprise.
By inputting macro factor data and historical load data of the generator set into the generator set load prediction model, the generator set load prediction value is obtained; then, these data and coal quality data of multiple coal mixing doping schemes are input to the raw coal consumption rate and factory power prediction model, and the raw coal consumption rate and factory power consumption power of each coal mixing doping scheme are predicted; based on these prediction values, the carbon emissions of the optimal coal mixing doping scheme are determined as the carbon emission prediction value.
By comprehensively considering macro factors, generator set equipment working conditions and coal-fired quality, the accuracy of carbon emission forecasts is improved, and the production optimization of power generation enterprises can be more effectively guided.
Smart Images
Figure CN117829345B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of data processing, and in particular to a carbon emission prediction method, a carbon emission prediction device and an electronic device. Background Art
[0002] When a power generation company has a surplus of carbon quota minus carbon emissions, it can sell the excess carbon quota for profit; conversely, when the company's carbon quota is less than its carbon emissions, the shortfall needs to be purchased from the carbon emissions trading market. Therefore, carbon emissions will become an important factor affecting the company's power generation costs. Predicting carbon emissions is of great significance for power generation companies to optimize production management and improve profitability.
[0003] At present, the main methods or models used for carbon emission prediction include IPAT model, STIRPAT model, logarithmic mean Dirichlet index method (LMDI), generalized Dirichlet index decomposition method (GDIM), LEAP, etc.
[0004] However, the above methods only consider macro factors to predict the total carbon emissions of a region, and do not predict carbon emissions based on the actual situation of the power generation company's equipment, resulting in low accuracy in carbon emissions prediction and no guidance for the production optimization of power generation companies. Summary of the invention
[0005] The purpose of the embodiments of the present invention is to provide a carbon emissions prediction method, a carbon emissions prediction device and an electronic device to solve the defect that the existing method only considers macro factors to predict the total carbon emissions of a region, does not predict carbon emissions based on the actual situation of the power generation enterprise's equipment, and leads to low accuracy of carbon emissions prediction.
[0006] In order to achieve the above object, an embodiment of the present invention provides a carbon emission prediction method, comprising:
[0007] Inputting the macro factor data within a future set time into the generator set load prediction model to obtain the generator set load prediction value within the future set time output by the generator set load prediction model;
[0008] Acquire multiple coal blending and combustion schemes and coal quality data corresponding to each of the coal blending and combustion schemes;
[0009] Input the predicted load value of the generator set and the coal quality data corresponding to all the coal blending and combustion schemes into the raw coal consumption rate and plant power prediction model, and obtain the raw coal consumption rate prediction value and plant power prediction value corresponding to each of the coal blending and combustion schemes within a future set time output by the raw coal consumption rate and plant power prediction model;
[0010] Determine the evaluation function value corresponding to each of the coal blending and combustion schemes based on the raw coal consumption rate prediction value and the plant power prediction value corresponding to each of the coal blending and combustion schemes;
[0011] Based on the evaluation function value corresponding to each of the coal blending and combustion schemes, determining the carbon emissions corresponding to the optimal coal blending and combustion scheme as the carbon emissions prediction value;
[0012] The generator set load prediction model is trained based on the historical macro-factor data and the historical load data of the generator set within the past set time, and the raw coal consumption rate and plant power prediction model is trained based on the historical load data of the generator set within the past set time, the historical coal quality data of the raw coal, the historical plant power data and the historical consumption rate data of the raw coal.
[0013] Optionally, determining the evaluation function value corresponding to each coal blending and combustion scheme based on the raw coal consumption rate prediction value and the plant power prediction value corresponding to each coal blending and combustion scheme includes:
[0014] Based on the coal quality data and raw coal consumption rate prediction value corresponding to each coal blending and combustion scheme, the carbon emissions corresponding to each coal blending and combustion scheme are calculated;
[0015] Based on the predicted value of raw coal consumption rate, predicted value of plant power consumption and carbon emissions corresponding to each of the coal blending and combustion schemes, the evaluation function value corresponding to each of the coal blending and combustion schemes is determined.
[0016] Optionally, determining the evaluation function value corresponding to each coal blending and combustion scheme based on the raw coal consumption rate prediction value, plant power prediction value and carbon emissions corresponding to each coal blending and combustion scheme includes:
[0017] Determine the raw coal consumption and plant power consumption corresponding to each of the coal blending and combustion schemes based on the raw coal consumption rate prediction value and the plant power prediction value corresponding to each of the coal blending and combustion schemes;
[0018] Based on the raw coal consumption, plant power consumption, carbon emissions, raw coal unit price, carbon emission rights unit price and power generation corresponding to each of the coal blending and combustion schemes, the evaluation function value corresponding to each of the coal blending and combustion schemes is determined.
[0019] Optionally, based on the raw coal consumption, power consumption, carbon emissions, raw coal unit price, carbon emission rights unit price and power generation corresponding to each of the coal blending and combustion schemes, the evaluation function value corresponding to each of the coal blending and combustion schemes is determined, which is expressed by the following formula:
[0020]
[0021] Among them, Cost is the evaluation function value of the coal blending and combustion scheme, K1 and K2 are adjustable parameters with a value range of [0,1], the raw coal consumption corresponding to the coal blending and combustion scheme includes the raw coal consumption of multiple coal bins, n is the total number of coal bins of the generator set, and the power generation is calculated based on the load forecast value of the generator set within the future set time.
[0022] Optionally, determining the carbon emissions corresponding to the optimal coal blending and combustion scheme as the carbon emissions prediction value based on the evaluation function value corresponding to each coal blending and combustion scheme includes:
[0023] The coal blending and combustion scheme with the smallest evaluation function value is determined as the optimal coal blending and combustion scheme, and the carbon emissions corresponding to the optimal coal blending and combustion scheme are determined as the carbon emissions prediction value.
[0024] Optionally, the macro factor data includes at least one of date, time, day of the week, holiday, climate and gross domestic product of the location of the generator set.
[0025] Optionally, the coal quality data includes at least one of a received basis lower calorific value, a received basis upper calorific value, moisture, ash, volatile matter, fixed carbon, carbon content, sulfur content and ash melting point.
[0026] Optionally, the generator set load prediction model is constructed based on a deep confidence network, and the raw coal consumption rate and plant power prediction model is constructed based on a deep confidence network.
[0027] On the other hand, an embodiment of the present invention further provides a carbon emission prediction device, comprising:
[0028] The first prediction module is used to input the macro factor data within a future set time into the generator set load prediction model to obtain the generator set load prediction value within the future set time output by the generator set load prediction model;
[0029] A coal quality data acquisition module, used to acquire multiple coal blending and combustion schemes and the coal quality data corresponding to each of the coal blending and combustion schemes;
[0030] The second prediction module is used to input the load prediction value of the generator set and the coal quality data corresponding to all the coal blending and combustion schemes into the raw coal consumption rate and plant power prediction model, and obtain the raw coal consumption rate prediction value and the plant power prediction value corresponding to each of the coal blending and combustion schemes within a future set time output by the raw coal consumption rate and plant power prediction model;
[0031] An evaluation function value determination module, used to determine the evaluation function value corresponding to each of the coal blending and combustion schemes based on the raw coal consumption rate prediction value and the plant power prediction value corresponding to each of the coal blending and combustion schemes;
[0032] A carbon emission prediction value calculation module is used to determine the carbon emission corresponding to the optimal coal blending and combustion scheme as the carbon emission prediction value based on the evaluation function value corresponding to each of the coal blending and combustion schemes;
[0033] The generator set load prediction model is trained based on the historical macro-factor data and the historical load data of the generator set within the past set time, and the raw coal consumption rate and plant power prediction model is trained based on the historical load data of the generator set within the past set time, the historical coal quality data of the raw coal, the historical plant power data and the historical consumption rate data of the raw coal.
[0034] On the other hand, the present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned carbon emission prediction method when executing the program.
[0035] On the other hand, the present invention also provides a machine-readable storage medium having a computer program stored thereon, and the computer program implements the above-mentioned carbon emission prediction method when executed by a processor.
[0036] Through the above technical scheme, the present invention predicts the raw coal consumption rate prediction value and the plant power prediction value corresponding to each of the coal blending and burning schemes through an artificial intelligence model, and then determines the evaluation function value corresponding to each of the coal blending and burning schemes based on the raw coal consumption rate prediction value, the plant power prediction value and the carbon emissions corresponding to each of the coal blending and burning schemes, and determines the carbon emissions corresponding to the optimal coal blending and burning scheme from the evaluation function values corresponding to each of the coal blending and burning schemes as the carbon emissions prediction value. Thus, the present invention comprehensively considers the macro factor data, the equipment operating conditions of the power plant generator set (raw coal consumption rate prediction value and plant power prediction value) and the quality of coal (evaluation function value), predicts carbon emissions, and improves the accuracy of carbon emissions prediction.
[0037] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings:
[0039] Figure 1It is a flow chart of the carbon emission prediction method provided by the present invention;
[0040] Figure 2 It is a structural schematic diagram of a generator set load prediction model constructed based on a deep belief network provided by the present invention;
[0041] Figure 3 It is a schematic diagram of the internal structure of the deep belief network provided by the present invention;
[0042] Figure 4 It is a structural schematic diagram of a raw coal consumption rate and plant power prediction model based on a deep belief network provided by the present invention;
[0043] Figure 5 It is a structural schematic diagram of the carbon emission prediction device provided by the present invention;
[0044] Figure 6 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION
[0045] The specific implementation of the embodiment of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the embodiment of the present invention, and is not used to limit the embodiment of the present invention.
[0046] Method Embodiment
[0047] Please refer to Figure 1 , an embodiment of the present invention provides a carbon emission prediction method, comprising:
[0048] Step 100: Input the macro factor data within a future set time into the generator set load prediction model to obtain the generator set load prediction value within the future set time output by the generator set load prediction model.
[0049] For example, the electronic device inputs the macro factor data within the next week into the generator set load prediction model, and obtains the generator set load prediction value within the next week output by the generator set load prediction model. Among them, the macro factor data includes at least one of the date, time, week, holiday, climate and gross domestic product of the generator set location. In order to comprehensively consider multiple factors and further improve the load prediction accuracy of the generator set load prediction model, in one embodiment, the macro factor data includes the date, time, week, holiday, climate and gross domestic product (GDP) of the generator set location. For example, the macro factor data can be the date 2010.1.1, the time 15:00, Monday, New Year's Day, sunny, and the gross domestic product of the generator set location is 2 billion yuan. Among them, the holiday can be represented by a numerical value, for example, it is 1 for a holiday and 0 for a non-holiday. The climate can also be represented by a numerical value, for example, it is 1 for sunny weather, 2 for cloudy weather, 3 for rainy weather, etc.
[0050] The generator set load prediction model can be constructed using various artificial intelligence models, such as convolutional neural networks. The generator set load prediction model is trained based on historical macro-factor data and historical load data of the generator set within a past set time. That is, the electronic device uses the historical macro-factor data within the past set time as sample data and the historical load data of the generator set as label data corresponding to the sample data to train the generator set load prediction model to obtain a trained generator set load prediction model. Thus, the electronic device inputs the macro-factor data within a future set time into the generator set load prediction model to obtain the generator set load prediction value within the future set time output by the generator set load prediction model. Among them, the training method of the generator set load prediction model can select various existing model training methods, but in order to avoid limiting this patent, no specific examples are given here.
[0051] Step 200: Acquire multiple coal blending and combustion schemes and coal quality data corresponding to each of the coal blending and combustion schemes.
[0052] Based on the current coal storage data of the power plant, the electronic device enumerates all reasonable coal blending and combustion schemes, obtains multiple coal blending and combustion schemes and the coal quality data corresponding to each of the coal blending and combustion schemes. In one embodiment, each of the coal blending and combustion schemes includes the coal quality data of multiple coal bunkers. For example, if a power plant has coal bunker A, coal bunker B, coal bunker C, coal bunker D, coal bunker E and coal bunker F, then each of the coal blending and combustion schemes includes the coal quality data of coal bunker A, coal quality data of coal bunker B, coal quality data of coal bunker C, coal quality data of coal bunker D, coal quality data of coal bunker E and coal quality data of coal bunker F.
[0053] Furthermore, the coal quality data of each coal bunker includes at least one of the received basis lower calorific value, received basis upper calorific value, moisture, ash, volatile matter, fixed carbon, carbon content, sulfur content, and ash melting point. Similarly, in order to comprehensively consider multiple factors and further improve the load prediction accuracy of the raw coal consumption rate and the plant power prediction model, in one embodiment, the coal quality data of each coal bunker includes the received basis lower calorific value, received basis upper calorific value, moisture, ash, volatile matter, fixed carbon, carbon content, sulfur content, and ash melting point.
[0054] In addition, the reasonable coal blending scheme is based on the design indicators of the generator set. For example, the total calorific value of each coal bunker, the desulfurization capacity index, the blending restriction conditions of low ash melting point coal, etc. In order to achieve the rated power generation, the average calorific value of the coal quality in each coal bunker needs to be no less than the specified threshold so that the boiler can provide sufficient steam; in order to meet the emission indicators, the average sulfur content of the coal quality in each coal bunker needs to be no higher than the specified threshold. If it exceeds the specified threshold, it may exceed the desulfurization capacity of the unit, making the chimney emissions not meet environmental protection standards; in order to reduce boiler coking, the design requirements will stipulate that the number of low ash melting point coal added does not exceed the specified threshold. These thresholds need to be set according to the design indicators of the installed generator set. The blending scheme that meets the design indicators of the generator set is a reasonable blending scheme.
[0055] Step 300: Input the predicted load value of the generator set and the coal quality data corresponding to all the coal blending and combustion schemes into the raw coal consumption rate and plant power prediction model, and obtain the raw coal consumption rate prediction value and plant power prediction value corresponding to each of the coal blending and combustion schemes within a future set time output by the raw coal consumption rate and plant power prediction model.
[0056] The electronic device inputs the generator set load prediction value output by the generator set load prediction model in step 100 and the coal quality data corresponding to all the coal blending and combustion schemes into the raw coal consumption rate and plant power prediction model, and the raw coal consumption rate and plant power prediction model outputs the raw coal consumption rate prediction value and the plant power prediction value corresponding to each of the coal blending and combustion schemes within a set future time.
[0057] Among them, the raw coal consumption rate and plant power prediction model can be constructed using various artificial intelligence models, such as convolutional neural networks. The raw coal consumption rate and plant power prediction model is trained based on the historical load data of the generator set, the historical coal quality data of the raw coal, the historical plant power data and the historical raw coal consumption rate data within the past set time. That is, the electronic device uses the historical load data of the generator set and the historical coal quality data of the raw coal within the past set time as sample data, and the historical plant power data and the historical raw coal consumption rate data as label data corresponding to the sample data to train the raw coal consumption rate and plant power prediction model, and obtains the trained raw coal consumption rate and plant power prediction model. Thus, the electronic device inputs the predicted value of the load of the generator set and the coal quality data corresponding to all the coal blending and combustion schemes into the raw coal consumption rate and plant power prediction model, and obtains the raw coal consumption rate prediction value and plant power prediction value corresponding to each of the coal blending and combustion schemes within the future set time output by the raw coal consumption rate and plant power prediction model. Among them, the training method of the raw coal consumption rate and plant power prediction model can select various existing model training methods, but no specific examples are given here to avoid limiting this patent.
[0058] Step 400: Determine the evaluation function value corresponding to each of the coal blending and combustion schemes based on the raw coal consumption rate prediction value and the plant power prediction value corresponding to each of the coal blending and combustion schemes.
[0059] Step 500: Based on the evaluation function value corresponding to each of the coal blending and combustion schemes, determine the carbon emissions corresponding to the optimal coal blending and combustion scheme as the carbon emissions prediction value.
[0060] The electronic device can determine the evaluation function value corresponding to each of the coal blending and burning schemes based on the predicted value of the raw coal consumption rate and the predicted value of the plant power consumption corresponding to each of the coal blending and burning schemes. The evaluation function value can be used to evaluate the quality of the coal. From the evaluation function values corresponding to all the coal blending and burning schemes, the carbon emissions corresponding to the optimal coal blending and burning scheme (with the highest coal quality) are determined as the predicted value of carbon emissions. Among them, the carbon emissions of the power generation unit can be calculated based on the coal quality data of the raw coal in the coal blending and burning scheme and the predicted raw coal consumption in accordance with the "Guidelines for Enterprise Greenhouse Gas Emissions Accounting and Reporting for Power Generation Facilities" issued by the Ministry of Ecology and Environment.
[0061] The present invention predicts the raw coal consumption rate prediction value and the plant power prediction value corresponding to each of the coal blending and burning schemes through an artificial intelligence model, and then determines the evaluation function value corresponding to each of the coal blending and burning schemes based on the raw coal consumption rate prediction value, the plant power prediction value and the carbon emissions corresponding to each of the coal blending and burning schemes, and determines the carbon emissions corresponding to the optimal coal blending and burning scheme from the evaluation function values corresponding to each of the coal blending and burning schemes as the carbon emissions prediction value. Thus, the present invention comprehensively considers the macro factor data, the equipment operating conditions of the power plant generator set (the raw coal consumption rate prediction value and the plant power prediction value) and the coal quality (evaluation function value), predicts the carbon emissions, and improves the accuracy of the carbon emissions prediction.
[0062] In other aspects of the embodiments of the present invention, step 400, based on the raw coal consumption rate prediction value and the plant power prediction value corresponding to each of the coal blending and combustion schemes, determines the evaluation function value corresponding to each of the coal blending and combustion schemes, including:
[0063] Step 410: Calculate the carbon emissions corresponding to each coal blending and combustion scheme based on the coal quality data corresponding to each coal blending and combustion scheme and the predicted value of the raw coal consumption rate.
[0064] The electronic device can integrate the predicted value of the raw coal consumption rate over a preset time (for example, 7 days) to obtain the raw coal consumption, and then calculate the carbon emissions of the power generation unit based on the coal quality data and raw coal consumption corresponding to each coal blending scheme in accordance with the "Guidelines for Enterprise Greenhouse Gas Emissions Accounting and Reporting for Power Generation Facilities" issued by the Ministry of Ecology and Environment.
[0065] Step 420: Determine the evaluation function value corresponding to each of the coal blending and combustion schemes based on the predicted value of the raw coal consumption rate, the predicted value of the plant power consumption, and the carbon emissions corresponding to each of the coal blending and combustion schemes.
[0066] The electronic device determines the evaluation function value corresponding to each of the coal blending and combustion schemes based on the raw coal consumption rate prediction value, the plant power prediction value, and the carbon emissions corresponding to each of the coal blending and combustion schemes, so as to facilitate determining the carbon emissions corresponding to the optimal coal blending and combustion scheme (with the highest coal quality) as the carbon emissions prediction value from the evaluation function values corresponding to all the coal blending and combustion schemes.
[0067] Further, step 420, based on the raw coal consumption rate prediction value, plant power prediction value and carbon emission corresponding to each of the coal blending and combustion schemes, determines the evaluation function value corresponding to each of the coal blending and combustion schemes, including:
[0068] Step 421: Determine the raw coal consumption and plant power consumption corresponding to each of the coal blending and combustion schemes based on the raw coal consumption rate prediction value and the plant power prediction value corresponding to each of the coal blending and combustion schemes.
[0069] The electronic device obtains the raw coal consumption amount by integrating the predicted value of the raw coal consumption rate corresponding to each of the coal blending and combustion schemes over a preset time, and obtains the plant power consumption by integrating the predicted value of the plant power corresponding to each of the coal blending and combustion schemes over a preset time.
[0070] Step 422: Determine the evaluation function value corresponding to each of the coal blending and combustion schemes based on the raw coal consumption, plant power consumption, carbon emissions, raw coal unit price, carbon emission rights unit price and power generation corresponding to each of the coal blending and combustion schemes.
[0071] The electronic device determines the evaluation function value corresponding to each of the coal blending and burning schemes based on the raw coal consumption, power consumption, carbon emissions, raw coal unit price, carbon emission rights unit price and power generation corresponding to each of the coal blending and burning schemes. The electronic device calculates the evaluation function values of all reasonable coal blending and burning schemes, selects the carbon emissions of the coal blending and burning scheme corresponding to the optimal evaluation function value, which is the final predicted carbon emissions. Therefore, the present invention comprehensively considers the macro factor data, the equipment operating conditions of the power plant generator set (the predicted value of the raw coal consumption rate and the predicted value of the power consumption) and the quality of the coal (evaluation function value), predicts the carbon emissions, and improves the accuracy of the carbon emission prediction.
[0072] Specifically, in one embodiment, step 422, based on the raw coal consumption, power consumption, carbon emissions, raw coal unit price, carbon emission rights unit price and power generation corresponding to each of the coal blending and combustion schemes, determines the evaluation function value corresponding to each of the coal blending and combustion schemes, which is expressed by the following formula:
[0073]
[0074] Among them, Cost is the evaluation function value of the coal blending and combustion scheme. The smaller the evaluation function value, the better the coal blending and combustion scheme. K1 and K2 are adjustable parameters with a value range of [0,1]. The raw coal consumption corresponding to the coal blending and combustion scheme includes the raw coal consumption of multiple coal bunkers. n is the total number of coal bunkers of the generator set. The power generation is based on the predicted load value of the generator set within the future set time and is calculated by integrating the preset time. Each of the coal blending and combustion schemes includes coal quality data of multiple coal bunkers. For example, a power plant has coal bunker A, coal bunker B, coal bunker C, coal bunker D, coal bunker E and coal bunker F, then each of the coal blending and combustion schemes includes coal quality data of coal bunker A, coal quality data of coal bunker B, coal quality data of coal bunker C, coal quality data of coal bunker D, coal quality data of coal bunker E and coal quality data of coal bunker F. Therefore, the raw coal consumption corresponding to the coal blending plan includes the raw coal consumption of coal bunker A, the raw coal consumption of coal bunker B, the raw coal consumption of coal bunker C, the raw coal consumption of coal bunker D, the raw coal consumption of coal bunker E and the raw coal consumption of coal bunker F.
[0075] In addition, K1 and K2 are adjustable parameters with a value range of [0,1]. When K1=1 and K2=0, it means that the evaluation function only considers the cost of raw coal and ignores the cost of carbon emission rights trading; when K1=0 and K2=1, it means that the evaluation function uses the reduction of carbon emissions as the judgment standard and ignores the cost of raw coal; when K1=1 and K2=1, it means that the evaluation function considers the cost of raw coal and the cost of carbon emission rights trading on the same importance. When the value range of K1 and K2 is (0,1), it means to find a balance between the cost of raw coal and the cost of carbon emission rights trading. In the embodiment of the present invention, the case where the value range of K1 and K2 is (0,1) can be selected.
[0076] When determining the evaluation function value corresponding to each of the coal blending and combustion schemes, step 500, based on the evaluation function value corresponding to each of the coal blending and combustion schemes, determines the carbon emissions corresponding to the optimal coal blending and combustion scheme as the carbon emissions prediction value, including: determining the coal blending and combustion scheme with the smallest evaluation function value as the optimal coal blending and combustion scheme, and determining the carbon emissions corresponding to the optimal coal blending and combustion scheme as the carbon emissions prediction value.
[0077] The electronic device can sort the evaluation function values corresponding to all coal blending and combustion schemes in ascending order, determine the coal blending and combustion scheme with the smallest evaluation function value as the optimal coal blending and combustion scheme, and determine the carbon emissions corresponding to the optimal coal blending and combustion scheme as the carbon emissions prediction value. Thus, the electronic device selects the coal blending and combustion scheme with the highest coal quality (smallest evaluation function value) from the evaluation function values corresponding to all coal blending and combustion schemes as the optimal coal blending and combustion scheme, and uses the carbon emissions corresponding to the optimal coal blending and combustion scheme as the carbon emissions prediction value to predict carbon emissions, thereby improving the accuracy of carbon emissions prediction.
[0078] In other aspects of the embodiments of the present invention, the generator set load prediction model is constructed based on a deep belief network. The model output result of the generator set load prediction model constructed based on the deep belief network is the load prediction value of the generator set within a future set time. For example, the structure of the deep belief network model is as follows: Figure 2 As shown in the figure, the future date, time, day of the week, holiday, climate and GDP of the location of the generator are input into the deep belief network (DBN network) to obtain the load forecast value of the generator. The Deep Belief Network (DBN) can be seen as a stack of a series of restricted Boltzmann machines (RBM). DBN belongs to a probabilistic graph model and is a mixture of directed and undirected graphs. Only the last two hidden layers are undirected graphs (this is an RBM), and the rest are directed graphs. The internal structure of the DBN network is as follows Figure 3 shown.
[0079] The following uses a three-layer DBN as an example to illustrate the network structure, which can be easily expanded to an N-layer DBN. Define θ = (w 0 , w 1 , b 0 , b 1 , b 2 ). The three-layer joint probability distribution is:
[0080] p(h 0 ,h 1 ,h 2 ;θ)=p(h 1 ,h 2 ;θ)p(h 0 |h 1 ;θ);
[0081] Since the last two hidden layers form an RBM:
[0082]
[0083] According to the conditional independence of the directed graph model, given h 1 The nodes in h 2 The nodes in are independent of each other (same as in RBM):
[0084]
[0085] Similarly, the raw coal consumption rate and plant power prediction model is constructed based on the deep belief network. The output result of the raw coal consumption rate and plant power prediction model constructed based on the deep belief network is the raw coal consumption rate and plant power of each coal bin of the generator set under the specified load and coal quality data. For example, the structure of the deep belief network model is as follows: Figure 4 As shown, the load of the generating set, the coal quality data of coal bunker A, the coal quality data of coal bunker B, the coal quality data of coal bunker C, the coal quality data of coal bunker D, the coal quality data of coal bunker E and the coal quality data of coal bunker F are input into the deep belief network (DBN network) to obtain the plant power, the raw coal consumption rate of coal bunker A, the raw coal consumption rate of coal bunker B, the raw coal consumption rate of coal bunker C, the raw coal consumption rate of coal bunker D, the raw coal consumption rate of coal bunker E and the raw coal consumption rate of coal bunker F.
[0086] In view of the fact that deep belief networks can process high-dimensional, large-scale, nonlinear data and have the characteristics of fast training speed and fast convergence speed, the embodiments of the present invention construct a generator set load prediction model and a raw coal consumption rate and plant power prediction model based on a deep belief network, which can improve the accuracy of load prediction values, raw coal consumption rate predictions of each coal bin, and plant power predictions, thereby improving the accuracy of carbon emissions predictions.
[0087] Device Embodiment
[0088] Please refer to Figure 5 On the other hand, the embodiment of the present invention also provides a carbon emission prediction device, including:
[0089] The first prediction module 501 is used to input the macro factor data within a future set time into the generator set load prediction model to obtain the generator set load prediction value within the future set time output by the generator set load prediction model;
[0090] A coal quality data acquisition module 502 is used to acquire a plurality of coal blending and combustion schemes and coal quality data corresponding to each of the coal blending and combustion schemes;
[0091] The second prediction module 503 is used to input the load prediction value of the generator set and the coal quality data corresponding to all the coal blending and combustion schemes into the raw coal consumption rate and plant power prediction model, and obtain the raw coal consumption rate prediction value and the plant power prediction value corresponding to each of the coal blending and combustion schemes within a future set time output by the raw coal consumption rate and plant power prediction model;
[0092] An evaluation function value determination module 504 is used to determine the evaluation function value corresponding to each of the coal blending and combustion schemes based on the raw coal consumption rate prediction value and the plant power prediction value corresponding to each of the coal blending and combustion schemes;
[0093] A carbon emission prediction value calculation module 505 is used to determine the carbon emission corresponding to the optimal coal blending and combustion scheme as the carbon emission prediction value based on the evaluation function value corresponding to each of the coal blending and combustion schemes;
[0094] The generator set load prediction model is trained based on the historical macro-factor data and the historical load data of the generator set within the past set time, and the raw coal consumption rate and plant power prediction model is trained based on the historical load data of the generator set within the past set time, the historical coal quality data of the raw coal, the historical plant power data and the historical consumption rate data of the raw coal.
[0095] Optionally, determining the evaluation function value corresponding to each coal blending and combustion scheme based on the raw coal consumption rate prediction value and the plant power prediction value corresponding to each coal blending and combustion scheme includes:
[0096] Based on the coal quality data and raw coal consumption rate prediction value corresponding to each coal blending and combustion scheme, the carbon emissions corresponding to each coal blending and combustion scheme are calculated;
[0097] Based on the predicted value of raw coal consumption rate, predicted value of plant power consumption and carbon emissions corresponding to each of the coal blending and combustion schemes, the evaluation function value corresponding to each of the coal blending and combustion schemes is determined.
[0098] Optionally, determining the evaluation function value corresponding to each coal blending and combustion scheme based on the raw coal consumption rate prediction value, plant power prediction value and carbon emissions corresponding to each coal blending and combustion scheme includes:
[0099] Determine the raw coal consumption and plant power consumption corresponding to each of the coal blending and combustion schemes based on the raw coal consumption rate prediction value and the plant power prediction value corresponding to each of the coal blending and combustion schemes;
[0100] Based on the raw coal consumption, plant power consumption, carbon emissions, raw coal unit price, carbon emission rights unit price and power generation corresponding to each of the coal blending and combustion schemes, the evaluation function value corresponding to each of the coal blending and combustion schemes is determined.
[0101] Optionally, based on the raw coal consumption, power consumption, carbon emissions, raw coal unit price, carbon emission rights unit price and power generation corresponding to each of the coal blending and combustion schemes, the evaluation function value corresponding to each of the coal blending and combustion schemes is determined, which is expressed by the following formula:
[0102]
[0103] Among them, Cost is the evaluation function value of the coal blending and combustion scheme, K1 and K2 are adjustable parameters with a value range of [0,1], the raw coal consumption corresponding to the coal blending and combustion scheme includes the raw coal consumption of multiple coal bins, n is the total number of coal bins of the generator set, and the power generation is calculated based on the load forecast value of the generator set within the future set time.
[0104] Optionally, determining the carbon emissions corresponding to the optimal coal blending and combustion scheme as the carbon emissions prediction value based on the evaluation function value corresponding to each coal blending and combustion scheme includes:
[0105] The coal blending and combustion scheme with the smallest evaluation function value is determined as the optimal coal blending and combustion scheme, and the carbon emissions corresponding to the optimal coal blending and combustion scheme are determined as the carbon emissions prediction value.
[0106] Optionally, the macro factor data includes at least one of date, time, day of the week, holiday, climate and gross domestic product of the location of the generator set.
[0107] Optionally, the coal quality data includes at least one of a received basis lower calorific value, a received basis upper calorific value, moisture, ash, volatile matter, fixed carbon, carbon content, sulfur content and ash melting point.
[0108] Optionally, the generator set load prediction model is constructed based on a deep confidence network, and the raw coal consumption rate and plant power prediction model is constructed based on a deep confidence network.
[0109] The carbon emission prediction device includes a processor and a memory. The above-mentioned first prediction module 501, coal quality data acquisition module 502, second prediction module 503, evaluation function value determination module 504 and carbon emission prediction value calculation module 505 are all stored in the memory as program units, and the processor executes the above-mentioned program units stored in the memory to realize corresponding functions.
[0110] The processor includes a kernel, which calls the corresponding program unit from the memory. There can be one or more kernels.
[0111] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.
[0112] Figure 6 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 6 As shown, the electronic device may include: a processor (processor) 610, a communication interface (Communications Interface) 620, a memory (memory) 630 and a communication bus 640, wherein the processor 610, the communication interface 620, and the memory 630 communicate with each other through the communication bus 640. The processor 610 can call the logic instructions in the memory 630 to execute the carbon emission prediction method, which includes: inputting the macro-factor data within a future set time into the generator set load prediction model to obtain the generator set load prediction value within the future set time output by the generator set load prediction model; obtaining multiple coal blending and combustion schemes and the coal quality data corresponding to each of the coal blending and combustion schemes; inputting the generator set load prediction value and the coal quality data corresponding to all the coal blending and combustion schemes into the raw coal consumption rate and plant power prediction model to obtain the raw coal consumption rate prediction value corresponding to each of the coal blending and combustion schemes within the future set time output by the raw coal consumption rate and plant power prediction model. and plant power prediction value; based on the raw coal consumption rate prediction value and the plant power prediction value corresponding to each of the coal blending and burning schemes, determine the evaluation function value corresponding to each of the coal blending and burning schemes; based on the evaluation function value corresponding to each of the coal blending and burning schemes, determine the carbon emissions corresponding to the optimal coal blending and burning scheme as the carbon emissions prediction value; the generator set load prediction model is trained based on the historical macro-factor data and the generator set historical load data within the past set time, and the raw coal consumption rate and plant power prediction model is trained based on the generator set historical load data, raw coal historical coal quality data, historical plant power data and raw coal historical consumption rate data within the past set time.
[0113] In addition, the logic instructions in the above-mentioned memory 630 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.
[0114] On the other hand, the present invention also provides a machine-readable storage medium having a computer program stored thereon, which is implemented when the processor executes the carbon emission prediction method, the method comprising: inputting macro-factor data within a future set time into a generator set load prediction model, and obtaining a generator set load prediction value within the future set time output by the generator set load prediction model; obtaining multiple coal blending and combustion schemes and coal quality data corresponding to each of the coal blending and combustion schemes; inputting the generator set load prediction value and the coal quality data corresponding to all the coal blending and combustion schemes into a raw coal consumption rate and plant power prediction model, and obtaining each of the coal blending and combustion schemes within the future set time output by the raw coal consumption rate and plant power prediction model. Corresponding raw coal consumption rate prediction value and plant power prediction value; based on the raw coal consumption rate prediction value and plant power prediction value corresponding to each of the coal blending and burning schemes, determine the evaluation function value corresponding to each of the coal blending and burning schemes; based on the evaluation function value corresponding to each of the coal blending and burning schemes, determine the carbon emissions corresponding to the optimal coal blending and burning scheme as the carbon emissions prediction value; the generator set load prediction model is trained based on the historical macro-factor data and the generator set historical load data within the past set time, and the raw coal consumption rate and plant power prediction model is trained based on the generator set historical load data, raw coal historical coal quality data, historical plant power data and raw coal historical consumption rate data within the past set time.
[0115] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.
[0116] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.
[0117] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A carbon emissions prediction method, characterized in that: include: Inputting the macro factor data within a future set time into the generator set load prediction model to obtain the generator set load prediction value within the future set time output by the generator set load prediction model; Acquire multiple coal blending and combustion schemes and coal quality data corresponding to each of the coal blending and combustion schemes; Input the predicted load value of the generator set and the coal quality data corresponding to all the coal blending and combustion schemes into the raw coal consumption rate and plant power prediction model, and obtain the raw coal consumption rate prediction value and plant power prediction value corresponding to each of the coal blending and combustion schemes within a future set time output by the raw coal consumption rate and plant power prediction model; Determine the evaluation function value corresponding to each of the coal blending and combustion schemes based on the raw coal consumption rate prediction value and the plant power prediction value corresponding to each of the coal blending and combustion schemes; Based on the evaluation function value corresponding to each of the coal blending and combustion schemes, determining the carbon emissions corresponding to the optimal coal blending and combustion scheme as the carbon emissions prediction value; The generator load prediction model is trained based on the historical macro factor data and the historical load data of the generator within the past set time, and the raw coal consumption rate and plant power prediction model is trained based on the historical load data of the generator within the past set time, the historical coal quality data of the raw coal, the historical plant power data and the historical consumption rate data of the raw coal; The macro-factor data include at least one of the date, time, day of the week, holidays, climate and the gross domestic product of the location of the generator set; the generator set load forecasting model is constructed based on a deep confidence network, and the raw coal consumption rate and plant power forecasting model is constructed based on a deep confidence network.
2. The carbon emission prediction method according to claim 1, characterized in that: Determining the evaluation function value corresponding to each coal blending and combustion scheme based on the raw coal consumption rate prediction value and the plant power prediction value corresponding to each coal blending and combustion scheme includes: Based on the coal quality data and raw coal consumption rate prediction value corresponding to each coal blending and combustion scheme, the carbon emissions corresponding to each coal blending and combustion scheme are calculated; Based on the predicted value of raw coal consumption rate, predicted value of plant power consumption and carbon emissions corresponding to each of the coal blending and combustion schemes, the evaluation function value corresponding to each of the coal blending and combustion schemes is determined.
3. The carbon emission prediction method according to claim 2, characterized in that: Determining the evaluation function value corresponding to each of the coal blending and combustion schemes based on the raw coal consumption rate prediction value, plant power prediction value, and carbon emissions corresponding to each of the coal blending and combustion schemes includes: Determine the raw coal consumption and plant power consumption corresponding to each of the coal blending and combustion schemes based on the raw coal consumption rate prediction value and the plant power prediction value corresponding to each of the coal blending and combustion schemes; Based on the raw coal consumption, plant power consumption, carbon emissions, raw coal unit price, carbon emission rights unit price and power generation corresponding to each of the coal blending and combustion schemes, the evaluation function value corresponding to each of the coal blending and combustion schemes is determined.
4. The carbon emission prediction method according to claim 3, characterized in that: Based on the raw coal consumption, power consumption, carbon emissions, raw coal unit price, carbon emission rights unit price and power generation corresponding to each of the coal blending and combustion schemes, the evaluation function value corresponding to each of the coal blending and combustion schemes is determined, which is expressed by the following formula: Among them, Cost is the evaluation function value of the coal blending and combustion scheme, K1 and K2 are adjustable parameters with a value range of [0,1], the raw coal consumption corresponding to the coal blending and combustion scheme includes the raw coal consumption of multiple coal bins, n is the total number of coal bins of the generator set, and the power generation is calculated based on the load forecast value of the generator set within the future set time.
5. The carbon emission prediction method according to claim 1, characterized in that: The step of determining the carbon emission amount corresponding to the optimal coal blending and combustion scheme as the carbon emission prediction value based on the evaluation function value corresponding to each coal blending and combustion scheme comprises: The coal blending and combustion scheme with the smallest evaluation function value is determined as the optimal coal blending and combustion scheme, and the carbon emissions corresponding to the optimal coal blending and combustion scheme are determined as the carbon emissions prediction value.
6. The carbon emission prediction method according to claim 1, characterized in that: The coal quality data includes at least one of the received basis lower calorific value, received basis upper calorific value, moisture, ash, volatile matter, fixed carbon, carbon content, sulfur content and ash melting point.
7. A carbon emission prediction device, characterized in that: include: The first prediction module is used to input the macro factor data within a future set time into the generator set load prediction model to obtain the generator set load prediction value within the future set time output by the generator set load prediction model; A coal quality data acquisition module, used to acquire multiple coal blending and combustion schemes and the coal quality data corresponding to each of the coal blending and combustion schemes; The second prediction module is used to input the load prediction value of the generator set and the coal quality data corresponding to all the coal blending and combustion schemes into the raw coal consumption rate and plant power prediction model, and obtain the raw coal consumption rate prediction value and the plant power prediction value corresponding to each of the coal blending and combustion schemes within a future set time output by the raw coal consumption rate and plant power prediction model; An evaluation function value determination module, used to determine the evaluation function value corresponding to each of the coal blending and combustion schemes based on the raw coal consumption rate prediction value and the plant power prediction value corresponding to each of the coal blending and combustion schemes; A carbon emission prediction value calculation module is used to determine the carbon emission corresponding to the optimal coal blending and combustion scheme as the carbon emission prediction value based on the evaluation function value corresponding to each of the coal blending and combustion schemes; The generator load prediction model is trained based on the historical macro factor data and the historical load data of the generator within the past set time, and the raw coal consumption rate and plant power prediction model is trained based on the historical load data of the generator within the past set time, the historical coal quality data of the raw coal, the historical plant power data and the historical consumption rate data of the raw coal; The macro-factor data include at least one of the date, time, day of the week, holidays, climate and the gross domestic product of the location of the generator set; the generator set load forecasting model is constructed based on a deep confidence network, and the raw coal consumption rate and plant power forecasting model is constructed based on a deep confidence network.
8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the carbon emission prediction method according to any one of claims 1 to 6 is implemented.
9. A machine-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the carbon emission prediction method according to any one of claims 1 to 6 is implemented.