A Decision Method for Reducing Carbon Costs of Regional Power Generation

Abstract

The present invention relates to a decision-making method for reducing the carbon cost of regional power generation. Based on real-time monitoring and cloud-edge collaboration, carbon emission data is captured through a continuous flue gas emission monitoring system to ensure the timeliness and reliability of the detection data, providing a reliable data basis for cloud platform prediction and decision-making. An enterprise-level edge computing gateway is used to achieve local information collection and data transmission continuation in case of network disconnection, ensuring information integrity and data reliability. By combining the distribution of edge computing and the centralization of the cloud platform, accurate prediction of the enterprise's carbon emission cost for the current year is achieved. Through cloud-edge collaboration, numerous carbon emission enterprises are connected into a whole, realizing interconnection and interoperability among enterprises, achieving comprehensive carbon emission benefits on a large scale, and having a very significant emission reduction effect.

Description

Technical Field

[0001] The present invention relates to a regional power generation carbon cost reduction decision-making method, and more specifically, to a regional power generation carbon cost reduction decision-making method based on real-time monitoring and cloud-edge collaboration, which is mainly used in regional low-carbon power dispatching departments. Background Art

[0002] The thermal power generation industry is an important source of carbon emissions. In a region, thermal power generation companies are usually geographically dispersed. In the existing technology, the emission factor method is mainly used to calculate carbon dioxide emissions, but it cannot meet the requirements of dynamic adjustment of carbon emissions in terms of timeliness and accuracy. At the same time, there are human errors, and it is impossible to achieve real-time dynamic adjustment of regional power generation carbon emissions, and it is impossible to make targeted emission reduction decisions. In addition, the current carbon emission control and management mechanism is mainly concentrated in a small range of a single thermal power company or several units. This small-scale management model greatly limits the improvement of emission reduction effects. Summary of the invention

[0003] The present invention is to avoid the shortcomings of the above-mentioned prior art and provide a regional power generation carbon cost reduction decision-making method, which predicts the annual carbon emission cost based on real-time carbon emission data in a larger area, and realizes dynamic adjustment of carbon emissions with the goal of maximizing power generation efficiency in a larger area, in order to obtain more significant emission reduction effects.

[0004] The present invention adopts the following technical solutions to achieve the purpose of the invention:

[0005] The regional power generation carbon cost reduction decision-making method of the present invention is characterized by being based on real-time monitoring and cloud-edge collaboration in the following steps:

[0006] Step 1. System settings:

[0007] A carbon emission real-time monitoring device and a cloud-edge collaborative device connected to a cloud platform server are set up for carbon emission enterprises included in carbon trading in the region; the carbon emission real-time monitoring device is used to obtain carbon emission real-time monitoring data; the cloud-edge collaborative device includes an enterprise-level edge computing gateway; a carbon emission cost calculation module, an enterprise production benefit optimization model, and a regional power generation plan optimization model are respectively set up in the cloud platform server;

[0008] Step 2: Data processing and transmission:

[0009] The enterprise-level edge computing gateway processes the real-time carbon emission monitoring data, removes shutdown and over-limit data, and obtains enterprise carbon emission pre-processing data, and the enterprise-level edge computing gateway transmits the enterprise carbon emission pre-processing data to the cloud platform server;

[0010] Step 3: Calculate the predicted carbon emission costs of each enterprise in the region this year

[0011] Divide this year into 6 cycles with a duration of 2 months each. Starting from the second cycle, calculate the predicted carbon emission costs of each enterprise in the region at the beginning of each cycle.

[0012] For enterprise A, at the start of the current cycle, use the carbon emission cost calculation module to calculate the carbon emissions E of enterprise A in the cycles that have occurred this year based on the preprocessed carbon emission data of enterprise A. A，n Extract the power generation Q of enterprise A in the cycles that have occurred this year from the cloud platform server. A，n The planned power generation F of enterprise A this year A The carbon quota Z of enterprise A this year A And the average carbon price P in the cycles that have occurred this year in the carbon trading market n Calculate the predicted carbon emission cost L of enterprise A this year. A Calculate the predicted carbon emission costs of each enterprise in the region in the same way; if the predicted carbon emission costs of all enterprises in the region this year are lower than the adjustment threshold value H, then go to step 7, otherwise go to step 4.

[0013] Step 4: Calculate the initial values of the power generation plan adjustment coefficients for each enterprise in the region:

[0014] For enterprise A, use the production benefit optimization model of enterprise A. The decision variable is the power generation plan adjustment coefficient, and the optimization goal is to maximize the production benefit of enterprise A. The value of the decision variable when the optimization goal is achieved is the initial value γ of the power generation plan adjustment coefficient of enterprise A. A Obtain the initial values of the power generation plan adjustment coefficients of each enterprise in the region in the same way and send them to the enterprise-level edge computing gateways of each enterprise correspondingly.

[0015] Step 5: Calculate the final values of the power generation plan adjustment coefficients for each enterprise in the region:

[0016] For each enterprise in the region that is determined to accept the power generation plan adjustment, use the regional power generation plan optimization model for overall regional optimization. The constraint condition is that the total power generation in the next cycle in the region meets the regional total power generation target. The decision variable is the final value of the power generation plan adjustment coefficient of each enterprise that accepts the power generation plan adjustment. The optimization goal is to maximize the sum TR of the production benefits of all enterprises that accept the power generation plan adjustment. The value of the decision variable when the optimization goal is achieved is the final value of the power generation plan adjustment coefficient of each enterprise.

[0017] Step 6: The cloud-edge collaborative device publishes the final values θ of the power generation plan adjustment coefficients to each enterprise that is determined to accept the power generation plan adjustment. B And implement the power generation plan adjustment within the current cycle until the end of the current cycle.

[0018] Step 7: Enter the next cycle. At the start of the next cycle, repeat Steps 1 - 6 until all 6 cycles of the whole year are completed, achieving the reduction of the regional annual carbon cost.

[0019] The characteristics of the regional power generation carbon cost reduction decision-making method of the present invention also lie in that Step 2 is implemented in the following manner:

[0020] Step 2.1: Use the continuous emission monitoring system CEMS installed at the chimney emission outlets of each thermal power unit in the regional thermal power plant to carry the carbon emission monitoring equipment to complete the installation of the carbon emission monitoring equipment.

[0021] Step 2.2: Connect the continuous emission monitoring system CEMS as a cloud-edge collaborative terminal to the data collection device of the edge computing gateway to realize the networking of CEMS, thereby realizing the transmission of the detection data obtained by the continuous emission monitoring system CEMS to the edge computing gateway, and at the same time transmitting the parameters of each thermal power unit in the regional thermal power plant to the edge computing gateway.

[0022] Step 2.3: Use the programmable logic controller in the edge computing gateway to perform network diagnosis, eliminate the downtime and over-limit data in the monitoring cycle, and obtain a preprocessed data set.

[0023] Let n represent the total number of monitoring windows in the occurred cycles, k be the monitoring window number, and the k-th window in the n monitoring windows is window k.

[0024] Let m represent the total number of thermal power units of enterprise A, j be the thermal power unit number, and the j-th unit in the m thermal power units is unit j.

[0025] The data in the preprocessed data set includes:

[0026] The running time t of unit j in window k j，k ;

[0027] The cross-sectional area S of the horizontal flue at the chimney inlet of unit j j ;

[0028] The average flue gas velocity V at the chimney inlet of unit j in window k j，k ;

[0029] The average pressure P of the horizontal flue at the chimney inlet of unit j in window k j，k ;

[0030] The average flue gas temperature T of the horizontal flue at the chimney inlet of unit j in window k j，k ;

[0031] The average flue gas humidity X of the horizontal flue at the chimney inlet of unit j in window k j，k ;

[0032] Average volume fraction of CO2 in the flue gas of unit j at window k

[0033] where k = 1, 2, … n; j = 1, 2, … m;

[0034] Step 2.4: Transmit the preprocessed data set to the cloud: The cloud platform uses a message middleware to set up a real-time data transmission service, build a data channel between the edge computing center and the central cloud, and each enterprise's edge computing gateway uploads the preprocessed data set in its monitoring cycle to the local cloud data center through the data channel.

[0035] The characteristics of the regional power generation carbon cost reduction decision-making method of the present invention also lie in:

[0036] The said step 3 calculates the predicted carbon emission cost of the enterprise in the current year as follows:

[0037] Step 3.1: The cloud platform server calculates according to the carbon emission preprocessing data of enterprise A in the following manner:

[0038] The dry flue gas exhaust volume Q of unit j at window k is calculated by formula (1) j，k as:

[0039]

[0040] The CO2 emission G of unit j at window k is calculated by formula (2) j，k as:

[0041]

[0042] Step 3.2: Calculate the total carbon emission of a single enterprise:

[0043] The carbon emission E of enterprise A in the occurred cycle in the current year is calculated by formula (3) A，n as:

[0044]

[0045] The predicted carbon emission cost L of enterprise A in the current year is calculated by formula (4) A as:

[0046] L A =(E A，n / Q A，n ×F A -Z A )×P n (4)

[0047] The predicted carbon emission costs of each enterprise in the region in the current year are calculated by the same method.

[0048] The characteristics of the decision-making method for reducing the carbon cost of regional power generation in the present invention also lie in that:

[0049] In step 4, the production benefit optimization model of enterprise A is constructed as formula (5):

[0050] Max(R A )=(P A，n -C A，n )×[Q A，n +(F A -Q A，n )×γ A

[0051] -[E A，n +E A，n / Q A，n ×(F A -Q A，n )×γ A -Z A ×P n (5)

[0052] In formula (5):

[0053] R A is the production benefit of enterprise A in the current year;

[0054] P A，n is the average electricity price of the cycle that has occurred in enterprise A in the current year;

[0055] C A，n is the average power generation cost of the cycle that has occurred in enterprise A in the current year;

[0056] Set the constraint condition as formula (6):

[0057] γ min ≤γ A ≤γ max (6)

[0058] Take γ A with a step of 0.01 and γ min as the initial value, vary between γ min and γ max , and use the exhaustive method to make the γ A value when R A takes the maximum value as the initial value of the power generation plan adjustment coefficient of enterprise A;

[0059] The characteristics of the decision-making method for reducing the carbon cost of regional power generation in the present invention also lie in that:

[0060] Step 5 is to obtain the final value of the power generation plan adjustment coefficient of each enterprise in the region according to the following method:

[0061] ​The cloud platform server performs overall optimization on all s enterprises in the region that are determined to accept the adjustment of the power generation plan, and while meeting the regional power generation target, maximizes the total net revenue TR of the power generation plan adjustment of the s enterprises; accordingly, the objective function expression included in the regional power generation plan optimization model is as shown in Equation (7):

[0062]

[0063] In Equation (7):

[0064] Let B represent the enterprise with the code B among the s enterprises, that is, enterprise B, where B = 1, 2,..., s;

[0065] θ B is the final value of the power generation plan adjustment coefficient of enterprise B;

[0066] TR is the total net revenue of the s enterprises;

[0067] P B，n is the average electricity price of the periods that have occurred in this year for enterprise B;

[0068] C B，n is the average power generation cost of the periods that have occurred in this year for enterprise B;

[0069] Q B，n is the power generation amount of the periods that have occurred in this year for enterprise B;

[0070] F B is the planned power generation amount of this year for enterprise B;

[0071] E B，n is the carbon emission amount of the periods that have occurred in this year for enterprise B;

[0072] Z B is the carbon quota of this year for enterprise B;

[0073] Set the constraint conditions as shown in Equation (8) and Equation (9):

[0074] θ min ≤θ B ≤θ max (8)

[0075]

[0076] Let θ B vary in steps of 0.01, with θ min as the initial value, between θ min and θ max Use the exhaustive method to obtain the value range of TR when θ B varies, and from the maximum value TR maxDetermine the final values of the power generation plan adjustment coefficients θ1, θ2, …, θ for all enterprises s ; F min is the minimum value of the planned power generation for all s enterprises set by the system for this year.

[0077] Compared with the prior art, the beneficial effects of the present invention are embodied in:

[0078] 1. The present invention timely captures carbon emission-related data through the continuous emission monitoring system CEMS, avoids interference from human factors, effectively ensures the timeliness and reliability of the detection data, and provides a reliable data basis for cloud platform prediction and decision-making.

[0079] 2. The present invention uses an enterprise-level edge computing gateway to achieve local collection of information and data transmission during network disconnection, effectively ensuring information integrity and data reliability. Prediction is achieved with the help of the cloud platform. Through the combination of the distributed edge computing and the centralized cloud platform, accurate prediction of the carbon emission cost of enterprises for this year is realized, and then the power generation adjustment coefficient for maximizing the power generation benefit of enterprises is calculated and achieved.

[0080] 3. The present invention connects numerous carbon emission enterprises into a whole through cloud-edge collaboration, realizes interconnection and intercommunication between enterprises, achieves comprehensive carbon emission benefits on a large scale, and the emission reduction effect is very significant. Description of the Drawings

[0081] Figure 1 is a schematic diagram of the logical relationship between the thermal power plant and cloud-edge collaboration in the present invention; Detailed Embodiment

[0082] The method for making a decision on reducing the carbon cost of regional power generation in the present invention is carried out based on real-time monitoring and cloud-edge collaboration according to the following steps:

[0083] Step 1. System setting:

[0084] Refer to Figure 1 , set carbon emission real-time monitoring devices and cloud-edge collaboration devices connected to the cloud platform server for carbon emission enterprises included in the carbon trading within the region; the carbon emission real-time monitoring devices are used to obtain carbon emission real-time monitoring data; the cloud-edge collaboration devices include enterprise-level edge computing gateways; a carbon emission cost calculation module, an enterprise production benefit optimization model, and a regional power generation plan optimization model are respectively set in the cloud platform server;

[0085] Step 2. Data processing and transmission:

[0086] The enterprise-level edge computing gateway processes the carbon emission real-time monitoring data, and after removing the shutdown and over-limit data, obtains the preprocessed enterprise carbon emission data, and the enterprise-level edge computing gateway transmits the preprocessed enterprise carbon emission data to the cloud platform server.

[0087] Step 3: Calculate the predicted carbon emission costs of each enterprise in the region for this year

[0088] Divide this year into 6 cycles with a duration of 2 months each. Starting from the second cycle, calculate the predicted carbon emission costs of each enterprise in the region at the beginning of each cycle;

[0089] The compliance cycle of the carbon emission trading market is usually from January 1st to December 31st of each year. Therefore, the reduction of regional power generation carbon costs needs to be calculated and implemented within the compliance cycle. Considering the seasonality and flexibility of the power generation plans of power generation enterprises, multiple power generation plan adjustments can be made according to market demand and enterprise benefits within a compliance cycle; To facilitate popularization and implementation, adapt to temperature changes, and avoid overly frequent adjustments, it is determined that each 2 months is a adjustment cycle.

[0090] The first cycle of power generation plan adjustment corresponds to the first 2 months of the carbon emission trading compliance cycle, and so on. The sixth cycle of power generation plan adjustment corresponds to the last 2 months of the carbon emission trading compliance cycle. When the sixth adjustment cycle ends, a new year and a new compliance cycle begin. At this time, it is necessary to statistically verify the carbon emissions of the previous compliance cycle. Therefore, the first cycle of power generation plan adjustment each year lacks data for this compliance cycle and there is no pressure to reduce carbon emissions. Therefore, no adjustment is made, and only relevant data needs to be collected to prepare for the second adjustment cycle.

[0091] For enterprise A, at the beginning of the current cycle, use the carbon emission cost calculation module to calculate the carbon emissions E of enterprise A in the cycles that have occurred this year based on the preprocessed carbon emission data of enterprise A A，n , extract the power generation Q of enterprise A in the cycles that have occurred this year from the cloud platform server A，n , the planned power generation F of enterprise A this year A , the carbon quota Z of enterprise A this year A , and the average carbon price P of the carbon trading market in the cycles that have occurred this year n , calculate the predicted carbon emission cost L of enterprise A this year A ; Calculate the predicted carbon emission costs of each enterprise in the region in the same way; If the predicted carbon emission costs of all enterprises in the region this year are lower than the adjustment threshold value H, then go to step 7, otherwise go to step 4.

[0092] Step 4: Calculate the initial values of the power generation plan adjustment coefficients of each enterprise in the region:

[0093] For Enterprise A, using the production benefit optimization model of Enterprise A, the decision variable is the power generation plan adjustment coefficient, and the optimization goal is to maximize the production benefit of Enterprise A. The value of the decision variable when the optimization goal is achieved is the initial value γ of the power generation plan adjustment coefficient of Enterprise A A , and the initial values of the power generation plan adjustment coefficients of each enterprise in the region are obtained by the same method and sent to the enterprise-level edge computing gateways of each enterprise correspondingly.

[0094] Step 5: Calculate the final values of the power generation plan adjustment coefficients of each enterprise in the region:

[0095] For each enterprise in the region that is determined to accept the power generation plan adjustment, use the regional power generation plan optimization model for overall regional optimization. The constraint condition is that the total power generation in the next cycle in the region meets the regional total power generation target. The decision variable is the final value of the power generation plan adjustment coefficient of each enterprise that accepts the power generation plan adjustment. The optimization goal is to maximize the sum TR of the production benefits of all enterprises that accept the power generation plan adjustment. The value of the decision variable when the optimization goal is achieved is the final value of the power generation plan adjustment coefficient of each enterprise;

[0096] In Step 4, the initial values of the power generation plan adjustment coefficients of each enterprise in the region are obtained and sent to the enterprise-level edge computing gateways of each enterprise correspondingly. At this time, some enterprises will make judgments based on their own unit conditions, enterprise cost-benefit conditions, and carbon emission rights trading conditions and will not accept the power generation plan adjustment coefficient, that is, they will not participate in the regional power generation carbon emission reduction decision-making in this adjustment cycle. Therefore, after the initial values of the power generation plan adjustment coefficients are sent to each enterprise, it is necessary to obtain the clear feedback of each enterprise to determine the scope of enterprises that accept the power generation plan adjustment. In Step 5, only the operations are performed on all enterprises in the region that are determined to accept the power generation plan adjustment, and the final values of the power generation plan adjustment coefficients of each enterprise are obtained.

[0097] Step 6: The cloud-edge collaborative device publishes the final value θ of the power generation plan adjustment coefficient to each enterprise that is determined to accept the power generation plan adjustment B ; and implement the power generation plan adjustment within the current cycle until the end of the current cycle.

[0098] Step 7: Enter the next cycle, and at the start of the next cycle, repeat Steps 1-6 until all 6 cycles of the whole year are completed to achieve regional annual carbon cost reduction.

[0099] In specific implementation, the corresponding technical measures also include:

[0100] Step 2 is implemented as follows:

[0101] Step 2.1: Use the continuous emission monitoring system CEMS installed at the chimney emission outlets of each thermal power unit in the regional thermal power plant to carry the carbon emission monitoring equipment to complete the installation of the carbon emission monitoring equipment;

[0102] Step 2.2: Connect the Continuous Emission Monitoring System (CEMS) as a cloud-edge collaboration terminal to the data collection device of the edge computing gateway to achieve the networking of CEMS, thereby realizing the transmission of detection data obtained from the CEMS to the edge computing gateway, and at the same time transmitting the parameters of each thermal power unit of the regional thermal power plants to the edge computing gateway;

[0103] CEMS is the abbreviation of English Continuous Emission Monitoting System, which refers to a device that continuously monitors the concentration and total emissions of gaseous pollutants and particulate matter discharged from atmospheric pollution sources and transmits the information to the competent department in real time. It is called the "flue gas automatic monitoring system", also known as the "continuous flue gas emission monitoring system" or "flue gas online monitoring system". CEMS is used for the real-time measurement of indicators such as flue gas flow rate and CO2 concentration in the flue gas of power generation enterprises, and uploads the data to the enterprise-level edge computing gateway and regional cloud platform through communication equipment to achieve the automation and informatization of data collection.

[0104] Step 2.3: Use the programmable logic controller in the edge computing gateway to perform network diagnosis, eliminate the downtime and overlimit data in the monitoring period, and obtain a preprocessed data set;

[0105] Let n represent the total number of monitoring windows in the occurred cycles, k be the monitoring window number, and the kth window among the n monitoring windows is window k;

[0106] Let m represent the total number of thermal power units of enterprise A, j be the thermal power unit number, and the jth unit among the m thermal power units is unit j;

[0107] The data in the preprocessed data set includes:

[0108] The operating time t of unit j in window k j，k ;

[0109] The cross-sectional area S of the horizontal flue at the chimney inlet of unit j j ;

[0110] The average flue gas flow velocity V of the horizontal flue at the chimney inlet of unit j in window k j，k ;

[0111] The average pressure P of the horizontal flue at the chimney inlet of unit j in window k j，k ;

[0112] The average flue gas temperature T of the horizontal flue at the chimney inlet of unit j in window k j，k ;

[0113] The average flue gas humidity X of the horizontal flue at the chimney inlet of unit j in window k j，k ;

[0114] The average volume fraction of CO2 in the flue gas of unit j at window k

[0115] where k = 1, 2, … n; j = 1, 2, … m;

[0116] Step 2.4: Transmit the pre - processed data set to the cloud: The cloud platform uses a message middleware to set up a real - time data transmission service, builds a data channel between the edge computing center and the central cloud, and each enterprise's edge computing gateway uploads the pre - processed data set in its enterprise's monitoring period to the cloud data center in its region through the data channel.

[0117] Step 3 is to calculate the predicted carbon emission cost of the enterprise in this year according to the following process:

[0118] Step 3.1: The cloud platform server calculates according to the carbon emission pre - processed data of enterprise A in the following way:

[0119] The dry flue gas exhaust volume Q of unit j at window k is calculated by formula (1) j，k as:

[0120]

[0121] The CO2 emission G of unit j at window k is calculated by formula (2) j，k as:

[0122]

[0123] Step 3.2: Calculate the total carbon emission of a single enterprise:

[0124] The carbon emission E of enterprise A in the occurred period of this year is calculated by formula (3) A，n as:

[0125]

[0126] The predicted carbon emission cost L of enterprise A in this year is calculated by formula (4) A as:

[0127] L A =(E A，n / Q A，n ×F A - Z A )×P n (4)

[0128] The predicted carbon emission costs of each enterprise in the region are calculated by the same method.

[0129] The production benefit optimization model of enterprise A constructed in Step 4 is formula (5):

[0130] Max(R A )=(P A，n -C A，n )×[Q A，n +(F A -Q A，n )×γ A

[0131] -[E A，n +E A，n / Q A，n ×(F A -Q A，n )×γ A -Z A ×P n (5)

[0132] In Equation (5):

[0133] R A is the production efficiency of Enterprise A in the current year;

[0134] P A，n is the average electricity price of the cycles that have occurred in the current year for Enterprise A;

[0135] C A，n is the average power generation cost of the cycles that have occurred in the current year for Enterprise A;

[0136] Set the constraint condition as Equation (6):

[0137] γ min ≤γ A ≤γ max (6)

[0138] Let γ A vary from γ min to γ min in steps of 0.01, with γ max as the initial value. Use the exhaustive method to find the value of γ A when R A takes the maximum value as the initial value of the power generation plan adjustment coefficient for Enterprise A;

[0139] Step 5 is to obtain the final values of the power generation plan adjustment coefficients of each enterprise in the region as follows:

[0140] The cloud platform server performs overall optimization on all s enterprises in the region that have determined to accept the power generation plan adjustment. While meeting the regional power generation target, maximize the total net revenue TR of the power generation plan adjustments of the s enterprises. Accordingly, construct the objective function expression included in the regional power generation plan optimization model as Equation (7):

[0141] ​

[0142] In Equation (7):

[0143] Let \(B\) represent the enterprise with enterprise code \(B\) among \(s\) enterprises, that is, enterprise \(B\), where \(B = 1, 2, \cdots, s\);

[0144] \(\theta\) B is the final value of the power generation plan adjustment coefficient of enterprise \(B\);

[0145] \(TR\) is the total net income of \(s\) enterprises;

[0146] \(P\) B，n is the average electricity price of the cycles that have occurred in the current year of enterprise \(B\);

[0147] \(C\) B，n is the average power generation cost of the cycles that have occurred in the current year of enterprise \(B\);

[0148] \(Q\) B，n is the power generation volume of the cycles that have occurred in the current year of enterprise \(B\);

[0149] \(F\) B is the planned power generation volume of enterprise \(B\) in the current year;

[0150] \(E\) B，n is the carbon emission volume of the cycles that have occurred in the current year of enterprise \(B\);

[0151] \(Z\) B is the carbon quota of enterprise \(B\) in the current year;

[0152] Set the constraint conditions as Equation (8) and Equation (9):

[0153] \(\theta\) min \(\leq \theta\) B \(\leq \theta\) max (8)

[0154]

[0155] Let \(\theta\) B change with a step of \(0.01\), with \(\theta\) min as the initial value, between \(\theta\) min and \(\theta\) max Use the exhaustive method to obtain the value range of \(TR\) when \(\theta\) B changes. Determine the final values of the power generation plan adjustment coefficients \(\theta_1, \theta_2, \cdots, \theta\) max of all enterprises from the maximum value \(TR\); s ;

[0156] \(F\) min is the minimum value of the planned power generation volume of all \(s\) enterprises set by the system in the current year.

[0157] Application example:

[0158] Six enterprises in a certain region are included in the carbon trading system. In each enterprise, carbon emission monitoring equipment is carried by the continuous emission monitoring system (CEMS) installed at the chimney emission outlets of each thermal power unit in the regional thermal power plant. An edge computing gateway is established in each enterprise in turn. At the same time, the edge computing gateway is connected to the cloud platform to build a data transmission channel. The year 2022 is selected as the research period. Carbon emission monitoring data is collected and uploaded starting from 00:01 on January 1, 2022, and the data collection within the first cycle is completed by 23:59 on February 28, 2022. The preprocessed data sets of each thermal power unit within the first cycle are obtained through collection and preprocessing by the edge computing gateway. The cloud platform receives the preprocessed data sets of the thermal power units and calculates the predicted carbon emission costs of each enterprise for this year by combining the average carbon price, the annual planned power generation of each enterprise, and the annual carbon quota index within the first cycle. Table 1 shows the parameters related to thermal power units and carbon emissions in each enterprise, including the total carbon emissions, the power generation already completed this year, and the planned annual power generation in the first cycle of 2022. The average carbon trading price is 58 yuan / ton, and the predicted carbon emission costs of the six enterprises for this year are calculated.

[0159] Table 1 Parameters related to thermal power unit equipment and carbon emissions in each enterprise

[0160]

[0161] Through prediction by the cloud platform server, the predicted annual carbon emission costs of each enterprise are obtained. It is found that the predicted values of the annual carbon emission costs of many enterprises are higher than the adjustment threshold value (set at 500,000 yuan). Then, the power generation enterprises in the region need to adjust their power generation plans. Based on data such as power generation costs and power generation revenue data, the enterprise production benefit optimization model is run, and the initial values of the power generation plan adjustment coefficients with the optimal benefits are calculated for each enterprise. The cloud platform sends the initial values of the power generation plan adjustment coefficients to each enterprise to solicit opinions on whether the enterprise accepts or does not accept the adjustment plan, and obtains clear feedback from each enterprise to determine the scope of enterprises that accept the power generation plan adjustment. For all enterprises in the region that are determined to accept the power generation plan adjustment, the regional emission benefit optimization model is run. After running, the final values of the power generation plan adjustment coefficients of each enterprise are obtained, and the final values of the power generation plan adjustment coefficients are sent to the edge computing gateways of each enterprise, which are executed by each enterprise in the second cycle.

[0162] Table 2 shows the relevant parameters of each enterprise in the second cycle. As shown in Table 2, after the six enterprises received the initial values of their respective power generation plan adjustment coefficients, they gave feedback to the cloud platform. Enterprises 1, 2, 3, 4, and 6 feedback that they accept the adjustment, while enterprise 5 does not accept the adjustment. Then, the five enterprises that accept the adjustment are considered for regional power generation plan optimization, and the final values of the power generation plan adjustment coefficients are obtained. These are sent by the cloud platform to the edge computing gateways of the five enterprises to achieve the optimal decision-making for reducing the regional power generation carbon cost.

[0163] Table 2 Calculation results of various parameters of each enterprise during this adjustment period

[0164]

[0165] By carrying out power generation adjustment on 5 thermal power enterprises in the region during the second power generation adjustment cycle, it is finally possible to increase the income of these 5 power generation enterprises by 232,000 yuan in the second cycle of this year, and reduce the operating costs brought by the carbon exchange.

Claims

1. A decision-making method for reducing the carbon cost of regional power generation, characterized in that Based on real-time monitoring and cloud-edge collaboration, it is carried out according to the following steps: Step 1. System setup: Set up real-time carbon emission monitoring devices for carbon emission enterprises included in the carbon trading within the region, as well as cloud-edge collaboration devices connected to the cloud platform server; the real-time carbon emission monitoring devices are used to obtain real-time carbon emission monitoring data; the cloud-edge collaboration devices include enterprise-level edge computing gateways; set up a carbon emission cost calculation module, an enterprise production benefit optimization model, and a regional power generation plan optimization model in the cloud platform server respectively; Step 2. Data processing and transmission: The enterprise-level edge computing gateway processes the real-time carbon emission monitoring data, and obtains enterprise carbon emission preprocessed data after removing shutdown and over-limit data, and the enterprise-level edge computing gateway transmits the enterprise carbon emission preprocessed data to the cloud platform server; Step 3. Calculate the predicted carbon emission cost of each enterprise in the region for this year Divide this year into 6 cycles with a duration of 2 months each. Starting from the second cycle, calculate the predicted carbon emission cost of each enterprise in the region for this year at the beginning of each cycle; For enterprise A, at the start of the current cycle, using the carbon emission cost calculation module, based on the preprocessed carbon emission data of enterprise A, calculate the carbon emission volume E of the cycles that have occurred in this year for enterprise A A，n , extract the generated electricity volume Q of the cycles that have occurred in this year for enterprise A from the cloud platform server A，n , the planned generated electricity volume F of enterprise A in this year A , the carbon quota Z of enterprise A in this year A , and the average carbon price P of the cycles that have occurred in this year in the carbon trading market n , calculate the predicted carbon emission cost L of enterprise A in this year A ; use the same method to calculate the predicted carbon emission costs of each enterprise in the region in this year; if the predicted carbon emission costs of all enterprises in the region in this year are lower than the adjustment threshold value H, then proceed to step 7, otherwise proceed to step 4; Step 4. Calculate the initial value of the power generation plan adjustment coefficient of each enterprise in the region: For Enterprise A, using the production benefit optimization model of Enterprise A, the decision variable is the power generation plan adjustment coefficient, and the optimization goal is to maximize the production benefit of Enterprise A. The value of the decision variable when the optimization goal is achieved is the initial value γ of the power generation plan adjustment coefficient of Enterprise A. A The initial values of the power generation plan adjustment coefficients of each enterprise in the region are obtained by the same method and sent to the enterprise-level edge computing gateways of each enterprise correspondingly. Step 5. Calculate the final value of the power generation plan adjustment coefficient of each enterprise in the region: For each enterprise in the region that is determined to accept the power generation plan adjustment, use the regional power generation plan optimization model for overall regional optimization. The constraint condition is that the total power generation in the next cycle within the region meets the regional total power generation target. The decision variable is the final value of the power generation plan adjustment coefficient of each enterprise that accepts the power generation plan adjustment. The optimization goal is to maximize the sum of production benefits TR of all enterprises that accept the power generation plan adjustment. The value of the decision variable when the optimization goal is achieved is the final value of the power generation plan adjustment coefficient of each enterprise; Step 6: The cloud-edge collaborative device publishes the final value θ of the power generation plan adjustment coefficient to each enterprise that has determined to accept the adjustment of the power generation plan B ; and implements the adjustment of the power generation plan within the current cycle until the end of the current cycle; Step 7. Enter the next cycle, and at the beginning of the next cycle, repeat steps 1-6 until all 6 cycles of the whole year are completed, so as to achieve regional annual carbon cost reduction.

2. The regional power generation carbon cost reduction decision method according to claim 1, characterized in that: Step 2 is implemented in the following manner; Step 2.

1. Use the continuous emission monitoring system CEMS installed at the chimney emission outlets of each thermal power unit in the regional thermal power plant to carry the carbon emission monitoring device to complete the installation of the carbon emission monitoring device; Step 2.

2. Connect the continuous emission monitoring system CEMS as a cloud-edge collaboration terminal to the data collection device of the edge computing gateway to realize the networking of CEMS, thereby realizing the transmission of the detection data obtained by the continuous emission monitoring system CEMS to the edge computing gateway, and at the same time transmitting the parameters of each thermal power unit in the regional thermal power plant to the edge computing gateway; Step 2.

3. Use the programmable logic controller in the edge computing gateway to perform network diagnosis, remove shutdown and over-limit data in the monitoring cycle, and obtain a preprocessed data set; Let n represent the total number of monitoring windows in the occurred cycles, k be the monitoring window number, and the kth window in the n monitoring windows is window k; Let m represent the total number of thermal power units of enterprise A, j be the thermal power unit number, and the jth unit in the m thermal power units is unit j; The data in the preprocessed data set includes: The operating time t of unit j at window k j，k ; The cross-sectional area S of the horizontal flue at the chimney inlet of unit j j ; The average flue gas velocity V at the chimney inlet of unit j in window k in the horizontal flue j，k ; Average pressure P of the horizontal flue at the chimney inlet of unit j at window k j，k ; The average flue gas temperature T at the chimney inlet of unit j in window k in the horizontal flue j，k ; The average humidity X of flue gas in the horizontal flue at the chimney inlet of unit j at window k j，k ; The average volume fraction of CO2 in the flue gas of unit j at window k where k = 1, 2, … n; j = 1, 2, … m; Step 2.4: Transmit the preprocessed data set to the cloud: The cloud platform uses a message middleware to set up a real-time data transmission service, build a data channel between the edge computing center and the central cloud, and each enterprise's edge computing gateway uploads the preprocessed data set in its enterprise's monitoring period to the local cloud data center through the data channel.

3. The regional power generation carbon cost reduction decision-making method according to claim 2, characterized in that: Step 3 is calculated to obtain the predicted carbon emission cost of the enterprise in this year as follows: Step 3.1: The cloud platform server calculates according to the carbon emission preprocessed data of enterprise A in the following manner: The dry flue gas exhaust volume Q of unit j in window k is calculated by formula (1). j，k It is as follows: The CO2 emissions G of unit j in window k are calculated by Equation (2). j，k It is as follows: Step 3.2: Calculate the total carbon emission of a single enterprise: The carbon emissions E of Enterprise A in the cycles that have occurred this year are calculated by Equation (3). A，n It is as follows: Calculate the predicted carbon emission cost L of Company A this year according to Equation (4). A It is as follows: L A = (E A，n / Q A，n × F A - Z A ) × P n (4) The predicted carbon emission costs of each enterprise in the region in this year are calculated by the same method.

4. The regional power generation carbon cost reduction decision-making method according to claim 1, characterized in that: The production benefit optimization model of enterprise A is constructed in step 4 as formula (5): Max(R A ) = (P A，n - C A，n ) × [Q A，n + (F A - Q A，n ) × γ A - A，n + E A，n / Q A，n × (F A - Q A，n ) × γ A - Z A × P n (5) In formula (5): R A is the production efficiency of Company A for this year; P A，n is the average electricity price of the cycles that have occurred in the current year for Enterprise A; C A，n is the average power generation cost of the occurred cycles of Enterprise A in this year; Set the constraint conditions as formula (6): γ min ≤γ A ≤γ max (6) Set γ A With a step size of 0.01, starting from γ min as the initial value, vary between γ min and γ max Using the exhaustive method, when R A takes the maximum value, the γ A value is the initial value of the power generation plan adjustment coefficient for Enterprise A.

5. The regional power generation carbon cost reduction decision-making method according to claim 1, characterized in that: Step 5 obtains the final value of the power generation plan adjustment coefficient of each enterprise in the region as follows: The cloud platform server performs overall optimization on all s enterprises in the region that are determined to accept the power generation plan adjustment. While meeting the regional power generation target, the total net income TR of the power generation plan adjustment of the s enterprises is maximized; accordingly, the objective function expression included in the regional power generation plan optimization model is constructed as formula (7): In formula (7): Let B represent the enterprise with the code B among the s enterprises, that is, enterprise B, B = 1, 2, … s; θ B is the final value of the power generation plan adjustment coefficient for Enterprise B; TR is the total net income of the s enterprises; P B，n is the average electricity price for the cycles that have occurred in the current year of Enterprise B; C B，n is the average power generation cost of the current period that has occurred for Company B this year; Q B，n is the generated electricity volume of the cycles that have occurred in this year for Enterprise B; F B is the planned power generation of Enterprise B for this year; E B，n is the carbon emissions of Enterprise B for the cycles that have occurred in this year; Z B is the carbon quota for Enterprise B in this year; Set the constraint conditions as formula (8) and formula (9): θ min ≤ θ B ≤ θ max (8) θ B With a step size of 0.01, min is the initial value, at θ min With θ max The exhaustive method is used to change between θ B The value range of TR is obtained when the maximum value TR max Determine the final values of the power generation plan adjustment coefficients θ1, θ2,…, θ s ; F min It is the minimum planned power generation of all s enterprises this year set by the system.

Images