Intelligent gas desulfurization system and method based on scheme optimization

Abstract

The invention relates to the technical field of gas desulfurization, in particular to an intelligent gas desulfurization system and method based on scheme optimization. The system comprises a data acquisition module, a data preprocessing module, a sulfur dioxide emission concentration intelligent prediction model, a gas pipe network optimization scheduling model and an output module. The prediction value of the sulfur dioxide emission concentration is calculated through registration of real-time data and historical data of the heating furnace device process and the gas pipe network system, and according to the prediction value of the sulfur dioxide emission concentration and the production-demand balance of the whole gas pipe network system, the standard emission of sulfur dioxide in the heating furnace flue gas is taken as the target. A scheduling scheme and a scheduling strategy of the gas pipe network system are adjusted and are displayed through a client, so that workers can carry out optimized scheduling operation conveniently, and energy conservation and emission reduction are realized; the problem that the emission concentration of sulfur dioxide in the process heating furnace tail gas of refinery enterprises exceeds the standard is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of gas desulfurization, and in particular to an intelligent system and method for gas desulfurization based on scheme optimization. Background Art

[0002] The gas pipeline network is an important part of the energy supply system of refineries and one of the main sources of fuel for refineries, providing fuel for the heating furnaces of various units in refineries. The gas pipeline network is the main equipment for transporting gas. The high-pressure gas directly generated by the operation of the refinery production unit will be directly incorporated into the high-pressure gas pipeline network and used as fuel for the heating furnaces of various units in the refinery. However, due to the high hydrogen sulfide content of this part of high-pressure gas and the poor quality of gas, the heating furnaces of various units are prone to exceed the sulfur dioxide emission concentration during operation.

[0003] On July 1, 2017, oil refineries began to strictly implement the atmospheric pollutant emission limits stipulated in the "Petroleum Refining Industry Pollutant Emission Standard" (GB31570-2015). Among them, the sulfur dioxide emission concentration in the flue gas emitted by the process heating furnace is more stringent. The sulfur dioxide emission concentration control of the process heating furnace has become one of the key tasks of flue gas pollutant control in refineries. In response to this situation, major refineries have conducted a lot of research, most of which are aimed at the optimization of the fuel gas system and the process flow of the gas desulfurization unit. This method requires adjustments and changes to the system structure of the device, which is costly and has a long cycle. There is a lack of scheduling optimization algorithms applied to the gas pipeline system. Summary of the invention

[0004] The technical problem to be solved by the present invention is: to overcome the shortcomings of the prior art, to provide an intelligent system and method for gas desulfurization based on scheme optimization, to dynamically adjust the gas pipe network system, to better predict the sulfur dioxide emission concentration of the heating furnace device and optimize the scheduling of the gas pipe network, to greatly improve the scheduling level of the gas pipe network and to reduce the sulfur dioxide emission concentration of the heating furnace device, and to reduce the problem of excessive sulfur dioxide emission concentration in the tail gas of the process heating furnace of the refinery enterprise.

[0005] The technical solution of the present invention is:

[0006] An intelligent system for gas desulfurization based on scheme optimization includes the following modules:

[0007] A data acquisition module is used to acquire data required for calculating the optimal scheduling plan and sulfur dioxide emission concentration from the gas pipe network system of the refinery device, and the data includes: real-time data of the heating furnace device process and the gas pipe network system, and historical data of the heating furnace device process and the gas pipe network system;

[0008] Data preprocessing module, used for preprocessing of process parameters, including:

[0009] Normal operating period data collection unit, used to extract data during normal operating period and delete data during shutdown state;

[0010] The abnormal data elimination unit is used to retain the collected product output data within the threshold range and eliminate the data outside the threshold range according to the preset data threshold range;

[0011] The data format unification unit is used to process the collected heating furnace device process and gas pipe network system process parameters in the same format, so as to facilitate the sulfur dioxide emission concentration intelligent prediction model to predict and output the sulfur dioxide emission concentration prediction value and the gas pipe network optimization scheduling model to provide a scheduling optimization plan;

[0012] An intelligent prediction model for sulfur dioxide emission concentration, which is used to predict the sulfur dioxide emission concentration of the heating furnace of the production device and obtain prediction data, including a model training unit and a model output unit;

[0013] The gas pipeline network optimization scheduling model includes:

[0014] A judgment unit, used to judge whether the hydrogen sulfide content of the gas in the gas pipeline used by the gas pipe network system to transport the gas to the excessive heating furnace is too high according to the sulfur dioxide emission concentration prediction data, and send the judgment result;

[0015] The scheduling optimization unit is used to optimize the scheduling strategy and scheduling plan of the gas system of each production device within the preset time period in the future according to the prediction data predicted by the prediction unit when receiving the result that the hydrogen sulfide content of a specific transportation route in the gas system of each production device within the preset time period in the future sent by the judgment unit is too high;

[0016] The output module is used to input the scheduling strategy and scheduling plan for optimizing the gas system of each production device within a preset time period in the future into the client, and display the optimized scheduling strategy and scheduling plan for the gas system of each production device within a preset time period in the future through the client, so that the scheduling personnel can perform optimized scheduling.

[0017] Preferably, the sulfur dioxide emission concentration intelligent prediction model adopts an RBF neural network model, and the construction of the RBF neural network model includes the following steps:

[0018] Step 1: In the RBF neural network structure, let X = [x1, x2, ..... xm] be the input vector of the neural network, and let B = [b1, b2, ..... bm] be the basis width variable of the neural network;

[0019] Step 2: The RBF neural network algorithm uses a Gaussian function as the radial basis function. The function expression is as follows:

[0020]

[0021] Where: x is the input variable; c j is the center of the Gaussian function; ||xc j || is the Euclidean norm; b is the base width parameter of the node;

[0022] Step 3: Calculation of sulfur dioxide emission concentration of heating furnace device:

[0023] The output value of the RBF neural network at time k is expressed as follows:

[0024] y m (k)=wh=w1h1+w2h2+.....+w m h m (2)

[0025] Where: w is the weight between the hidden layer and the output layer; ym(k) is the actual output;

[0026] Step 4: Output value verification:

[0027] When the output value y at time k m When the mean square error between the actual value y(k) at time k and the actual value y(k) at time k is less than the allowable error, it is proved that the output value is available. The calculation formula is as follows:

[0028] N(k)=1 / 2[y(k)-y m (k)] 2 (3).

[0029] Preferably, the model training unit of the sulfur dioxide emission concentration intelligent prediction model uses historical data to adjust the prediction accuracy of the model;

[0030] The model output module of the sulfur dioxide emission concentration intelligent prediction model outputs the predicted value of sulfur dioxide emission concentration using the trained prediction model.

[0031] Preferably, the gas network optimization scheduling model uses the predicted value of sulfur dioxide emission concentration, gas transportation equipment and pipelines to establish a gas network optimization scheduling model, construct a reward and punishment function for scheme evaluation under the current scheduling scheme, and use the predicted output sulfur dioxide emission concentration as the evaluation criterion.

[0032] Preferably, the gas pipeline network optimization scheduling model uses historical data to screen out consumer devices and treatment facilities with more exceeding-meter times, and modifies the gas scheduling plan based on the number of exceeding-meter times and the exceeding-meter concentration, giving priority to gas desulfurization treatment, and reintegrating the treated gas into the pipeline network, thereby reducing the sulfur dioxide emission concentration of the consumer devices.

[0033] Another technical solution of the present invention is:

[0034] A method for a gas desulfurization intelligent system based on scheme optimization comprises the following steps:

[0035] A. Collect real-time and historical data of heating furnace device process and gas pipe network system and perform data preprocessing;

[0036] B. Intelligent prediction model of sulfur dioxide emission concentration uses process parameters to predict the sulfur dioxide emission concentration of the output heating furnace;

[0037] C. Taking the predicted value of sulfur dioxide emission concentration of heating furnace as the judgment standard, construct the gas production device-gas pipeline network-gas consumption device gas pipeline network optimization scheduling model;

[0038] D. Output the optimized scheduling strategy and scheduling plan for the gas system of each production unit within the preset time period in the future.

[0039] Preferably, the data preprocessing described in step A includes: real-time acquisition of all point data, deleting data during downtime, and performing process parameter preprocessing;

[0040] The threshold value of the process parameters after cleaning in the previous step is calculated. The process parameters included in the model are uniformly calculated after being screened in the above steps, and a value range of each parameter is obtained. The value range is adjusted in combination with the properties of each process parameter and their actual rated working conditions. The value range of each process parameter is determined. The process parameters used for model construction are the process parameters after the value range is divided by the process parameter screening module;

[0041] The data preprocessing described in step A includes: performing key parameter analysis on the relevant process parameters processed in the previous step, and incorporating the mined key parameters into the model.

[0042] Preferably, the gas network optimization scheduling model described in step C determines the gas production and consumption devices and the gas transportation network according to the scheduling optimization process of the gas network system, determines the desulfurization treatment facilities and processes used by the gas consumption devices, and establishes a production-transportation-consumption and waste gas treatment facility gas network optimization scheduling model;

[0043] The reward and penalty function of the scheme evaluation under the current dispatching scheme is adopted, and the predicted value of sulfur dioxide emission concentration is used as the evaluation criterion;

[0044] According to the received sulfur dioxide emission concentration forecast value, determine the number of heating furnaces exceeding the standard in the current state;

[0045] According to the preset plan, adjust the gas dispatching plan of the heating furnace whose predicted sulfur dioxide emission concentration exceeds the standard;

[0046] Recalculate the data after the adjustment of the dispatching plan and predict the output sulfur dioxide emission concentration. If the predicted sulfur dioxide emission concentration is less than the limit, the gas dispatching plan will be transmitted to the client for display;

[0047] If the recalculated predicted value of sulfur dioxide emission concentration is still higher than the limit, the gas network optimization scheduling model will be returned to re-scheduling the plan until the predicted value of sulfur dioxide emission concentration is lower than the limit;

[0048] It is not activated when the predicted sulfur dioxide emission concentration is lower than the limit.

[0049] Preferably, the output of the optimized scheduling strategy and scheduling plan of the gas system of each production unit within a preset time period in the future described in step D is specifically for the state where the predicted value of sulfur dioxide emission concentration exceeds the standard under the current gas scheduling plan, and the gas pipeline network optimization scheduling model adjusts the gas scheduling plan. If the predicted value of sulfur dioxide emission concentration is lower than the limit value under this plan, the adjusted scheduling plan will be transmitted to the client port for display, so that the staff can perform optimized scheduling operations.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] By registering the real-time and historical data of the heating furnace device process and the gas pipe network system, the predicted value of sulfur dioxide emission concentration is calculated. Based on the predicted value of sulfur dioxide emission concentration and the production and demand balance of the entire gas pipe network system, the dispatching plan and dispatching strategy of the gas pipe network system are adjusted with the goal of achieving the standard emission of sulfur dioxide in the flue gas of the heating furnace. The results are displayed through the client, so that the staff can optimize the dispatching operation and achieve energy saving and emission reduction;

[0052] It can predict the dispatching plan and dispatching strategy of the output gas pipe network system without adding additional equipment or changing the existing gas pipe network structure, which is convenient for the staff to adjust and control in time and avoids excessive emission of sulfur dioxide in the flue gas of the heating furnace device. It has extremely high theoretical research value and practical application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0054] Figure 1 It is a flow diagram of the present invention. DETAILED DESCRIPTION

[0055] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0056] Embodiment 1

[0057] In this example, if Figure 1 As shown in the figure, the following steps are used to implement the dispatching and control of the gas pipeline network of the refinery enterprise based on the excessively high sulfur content of the gas:

[0058] Step A, determine the production plan interval and the devices for producing and consuming gas and the gas pipeline network system for transporting gas, and establish a gas pipeline network optimization scheduling model.

[0059] The parameters of the gas network pressure, production equipment and consumption equipment depend on the application example.

[0060] Step B, collecting real-time data and historical data of the heating furnace device process and gas pipe network system and performing data preprocessing to be included in the model data screening;

[0061] Step C, specifically, the real-time data and historical data process parameters of the heating furnace device process and the gas pipe network system include the amine solution concentration of the heating furnace treatment facility, the amine solution circulation volume of the heating furnace treatment facility, the washing water flow rate, the regeneration tower top flow rate, and the gas flow rate;

[0062] Step D, incorporating the above process parameters into the intelligent prediction model for sulfur dioxide emission concentration;

[0063] Preferably, but not limitedly, the sulfur dioxide emission concentration intelligent prediction model adopts a neural network algorithm, and the algorithm model may adopt an RBF artificial neural network;

[0064] Step 1: In the RBF neural network structure, let X = [x1, x2, ..... xm] be the input vector of the neural network, and let B = [b1, b2, ..... bm] be the basis width variable of the neural network;

[0065] Step 2: The RBF neural network algorithm uses a Gaussian function as the radial basis function. The function expression is as follows:

[0066]

[0067] Where: x is the input variable; c j is the center of the Gaussian function; ||xc j || is the Euclidean norm; b is the base width parameter of the node;

[0068] Step 3: Calculation of sulfur dioxide emission concentration of heating furnace device:

[0069] The output value of the RBF neural network at time k is expressed as follows:

[0070] y m (k)=wh=w1h1+w2h2+.....+w m h m (2)

[0071] Where: w is the weight between the hidden layer and the output layer; ym(k) is the actual output;

[0072] Step 4: Output value verification:

[0073] When the output value y at time k m When the mean square error between the actual value y(k) at time k and the actual value y(k) at time k is less than the allowable error, it is proved that the output value is available. The calculation formula is as follows:

[0074] N(k)=1 / 2[y(k)-y m (k)] 2 (3).

[0075] Step E: Transmitting the sulfur dioxide emission concentration intelligent prediction model to the gas network optimization scheduling model.

[0076] The gas network optimization scheduling model adopts the reward and penalty function of the scheme evaluation under the current scheduling scheme, and the predicted value of sulfur dioxide emission concentration is used as the evaluation criterion;

[0077] Step F, the gas network optimization scheduling model adjusts the gas scheduling plan of the heating furnace whose predicted sulfur dioxide emission concentration exceeds the standard according to the preset plan;

[0078] Step G, the sulfur dioxide emission concentration prediction model recalculates according to the data after the scheduling plan is adjusted, and predicts and outputs the sulfur dioxide emission concentration. If the predicted value of the sulfur dioxide emission concentration is less than the limit value, the gas scheduling plan is transmitted to the client for display;

[0079] Specifically, if the recalculated sulfur dioxide emission concentration prediction value of the sulfur dioxide emission concentration prediction model is still higher than the limit value, the gas pipe network optimization scheduling model is returned to re-scheduling the plan until the sulfur dioxide emission concentration prediction value is lower than the limit value;

[0080] Step H, transmitting the gas dispatching plan output by the gas pipeline network optimization dispatching model to the client for display;

[0081] Specifically, the gas dispatching scheme is a scheme for reducing the predicted value of the sulfur dioxide emission concentration intelligent prediction model to below the limit value;

[0082] Specifically, the gas network optimization scheduling model is activated only when the predicted value of the sulfur dioxide emission concentration intelligent prediction model is higher than the limit value.

[0083] Example:

[0084] XX refinery was selected for example experiment. The high-pressure gas pipeline pressure of the refinery is 0.5Mpa, and the low-pressure gas pipeline pressure is 0.1Mpa. The gas output of the refinery is 40,000t / a. The main production unit of gas is catalytic cracking unit, and a small part comes from polypropylene unit. The units that consume gas include catalytic waste heat boiler, catalytic atmospheric pressure boiler, heating furnace, etc.

[0085] Two catalytic waste heat boilers and three heating furnaces of the refinery were selected as experimental objects. The examples were named waste heat boiler 1#, waste heat boiler 2#, heating furnace 1#, heating furnace 2#, and heating furnace 3#. A three-day experiment was conducted. The sulfur dioxide emission concentration monitoring frequency was 3 minutes / time. The intelligent system and method for gas desulfurization based on scheme optimization proposed in this patent were adopted. The experimental results are shown below:

[0086] Table 1

[0087]

[0088] Two catalytic waste heat boilers and three heating furnaces of the refinery were selected as experimental objects, and the examples were named waste heat boiler 1#, waste heat boiler 2#, heating furnace 1#, heating furnace 2#, and heating furnace 3#. A three-day experiment was conducted, and the sulfur dioxide emission concentration monitoring frequency was 3 minutes / time. The intelligent system and method for gas desulfurization based on scheme optimization proposed in this patent was not adopted. The experimental results are shown below (the selected example device is the same as the device in Table 1):

[0089] Table 2

[0090]

[0091] As shown in the above diagram, when the intelligent system and method for gas desulfurization based on scheme optimization proposed in this patent are adopted, the sulfur dioxide emission concentration of the five experimental devices exceeded the standard three times, and the maximum exceeding rate was 0.21%. When the intelligent system and method for gas desulfurization based on scheme optimization proposed in this patent are not adopted, the sulfur dioxide emission concentration of the five experimental devices exceeded the standard ten times at most, and the maximum exceeding rate was 0.69%. The use of this patented technology greatly reduces the excess rate of sulfur dioxide emission concentration.

[0092] Embodiment three:

[0093] XX refinery was selected for example experiment. The high-pressure gas pipeline pressure of the refinery is 0.6Mpa, the low-pressure gas pipeline pressure is 0.1Mpa, the gas output of the refinery is 160,000t / a, the main gas production units include catalytic cracking unit, wax oil hydrogenation unit, coking unit, gas separation unit, olefin unit, polypropylene unit, etc., the gas consuming units include catalytic waste heat boiler, catalytic atmospheric pressure boiler, continuous reforming heating furnace, waste acid unit, coking unit, etc.

[0094] Two continuous reforming boilers, two catalytic atmospheric pressure boilers and two coking furnaces of the refinery were selected as experimental objects. The examples were named continuous reforming boiler 1#, continuous reforming boiler 2#, catalytic atmospheric pressure boiler 1#, catalytic atmospheric pressure boiler 2#, coking furnace 1# and coking furnace 2#. A three-day experiment was conducted. The sulfur dioxide emission concentration monitoring frequency was 3 minutes / time. An intelligent gas desulfurization system and method based on scheme optimization proposed in this patent was used. The experimental results are shown below:

[0095] Table 3

[0096]

[0097] Two continuous reforming boilers, two catalytic atmospheric pressure boilers and two coking furnaces of the refinery were selected as experimental objects. The examples were named continuous reforming boiler 1#, continuous reforming boiler 2#, catalytic atmospheric pressure boiler 1#, catalytic atmospheric pressure boiler 2#, coking furnace 1# and coking furnace 2#. A three-day experiment was conducted. The sulfur dioxide emission concentration monitoring frequency was 3 minutes / time. An intelligent system and method for gas desulfurization based on scheme optimization proposed in this patent was adopted. The experimental results are shown below (the selected example device is the same as the device in Table 3):

[0098] Table 4

[0099]

[0100] As shown in the above diagram, when the intelligent system and method for gas desulfurization based on scheme optimization proposed in this patent are adopted, the sulfur dioxide emission concentration of the six experimental devices exceeds the standard for a maximum of 3 times, and the maximum exceedance rate is 0.21%. When the intelligent system and method for gas desulfurization based on scheme optimization proposed in this patent are not adopted, the sulfur dioxide emission concentration of the five experimental devices exceeds the standard for a maximum of 10 times, and the maximum exceedance rate is 0.69%.

[0101] To sum up the above two examples, in the three-day experiment, the example in Example 1 exceeded the standard a total of 9 times when the patent technology was adopted, and the example in Example 2 exceeded the standard a total of 9 times when the patent technology was adopted. Without the use of the patent technology, the number of times the table was exceeded was 32 times and 48 times respectively. It can be concluded that when refineries adopt an intelligent system and method for gas desulfurization based on scheme optimization proposed in this patent, the number of sulfur dioxide emission concentrations exceeding the standard of consumption equipment using gas as fuel in refineries can be significantly reduced.

[0102] Although the present invention has been described in detail by referring to the accompanying drawings and in combination with the preferred embodiments, the present invention is not limited thereto. Without departing from the spirit and essence of the present invention, a person of ordinary skill in the art may make various equivalent modifications or substitutions to the embodiments of the present invention, and these modifications or substitutions shall be within the scope of the present invention. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, and all of these shall be within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be subject to the scope of protection of the claims.

Claims

1. An intelligent system for gas desulfurization based on scheme optimization, characterized in that: Includes the following modules: A data acquisition module is used to acquire data required for calculating the optimal scheduling plan and sulfur dioxide emission concentration from the gas pipe network system of the refinery device, and the data includes: real-time data of the heating furnace device process and the gas pipe network system, and historical data of the heating furnace device process and the gas pipe network system; Data preprocessing module, used for preprocessing of process parameters, including: Normal operating period data collection unit, used to extract data during normal operating period and delete data during shutdown state; The abnormal data elimination unit is used to retain the collected product output data within the threshold range and eliminate the data outside the threshold range according to the preset data threshold range; The data format unification unit is used to process the collected heating furnace device process and gas pipe network system process parameters in the same format, so as to facilitate the sulfur dioxide emission concentration intelligent prediction model to predict and output the sulfur dioxide emission concentration prediction value and the gas pipe network optimization scheduling model to provide a scheduling optimization plan; An intelligent prediction model for sulfur dioxide emission concentration, which is used to predict the sulfur dioxide emission concentration of the heating furnace of the production device and obtain prediction data, including a model training unit and a model output unit; The gas pipeline network optimization scheduling model includes: A judgment unit, used to judge whether the hydrogen sulfide content of the gas in the gas pipeline used by the gas pipe network system to transport the gas to the excessive heating furnace is too high according to the sulfur dioxide emission concentration prediction data, and send the judgment result; The scheduling optimization unit is used to optimize the scheduling strategy and scheduling plan of the gas system of each production device within the preset time period in the future according to the prediction data predicted by the prediction unit when receiving the result that the hydrogen sulfide content of a specific transportation route in the gas system of each production device within the preset time period in the future sent by the judgment unit is too high; The output module is used to input the scheduling strategy and scheduling plan for optimizing the gas system of each production device within a preset time period in the future into the client, and display the optimized scheduling strategy and scheduling plan for the gas system of each production device within a preset time period in the future through the client, so that the scheduling personnel can perform optimized scheduling.

2. The intelligent system for gas desulfurization based on scheme optimization according to claim 1 is characterized in that: The sulfur dioxide emission concentration intelligent prediction model adopts the RBF neural network model. The construction of the RBF neural network model includes the following steps: Step 1: In the RBF neural network structure, let X = [x1, x2, ..... xm] be the input vector of the neural network, and let B = [b1, b2, ..... bm] be the basis width variable of the neural network; Step 2: The RBF neural network algorithm uses a Gaussian function as the radial basis function. The function expression is as follows: Where: x is the input variable; c j is the center of the Gaussian function; ||xc j || is the Euclidean norm; b is the base width parameter of the node; Step 3: Calculation of sulfur dioxide emission concentration of heating furnace device: The output value of the RBF neural network at time k is expressed as follows: y m (k)=wh=w1h1+w2h2+.......+w m h m (2) Where: w is the weight between the hidden layer and the output layer; ym(k) is the actual output; Step 4: Output value verification: When the output value y at time k m When the mean square error between the actual value y(k) at time k and the actual value y(k) at time k is less than the allowable error, it is proved that the output value is available. The calculation formula is as follows: N(k)=1 / 2[y(k)-y m (k)] 2 (3)。 3. The intelligent system for gas desulfurization based on scheme optimization according to claim 1 is characterized in that: The model training unit of the sulfur dioxide emission concentration intelligent prediction model is to adjust the prediction accuracy of the model by using historical data; The model output module of the sulfur dioxide emission concentration intelligent prediction model outputs the predicted value of sulfur dioxide emission concentration using the trained prediction model.

4. The intelligent system for gas desulfurization based on scheme optimization according to claim 1 is characterized in that: The gas pipeline network optimization scheduling model uses the predicted value of sulfur dioxide emission concentration, gas transportation devices and pipelines to establish the gas pipeline network optimization scheduling model, constructs a reward and punishment function for program evaluation under the current scheduling plan, and uses the predicted output sulfur dioxide emission concentration as the evaluation criterion.

5. The intelligent system for gas desulfurization based on scheme optimization according to claim 1 is characterized in that: The gas pipeline network optimization scheduling model uses historical data to screen out consumer devices and treatment facilities with a large number of exceeding the meter limit, and modifies the gas scheduling plan based on the number of exceeding the standard and the exceeding meter concentration, giving priority to gas desulfurization treatment, and reintegrating the treated gas into the pipeline network, thereby reducing the sulfur dioxide emission concentration of the consumer devices.

6. A method of a gas desulfurization intelligent system based on scheme optimization, using the gas desulfurization intelligent system based on scheme optimization as claimed in any one of claims 1 to 5, characterized in that: The steps include: A. Collect real-time and historical data of heating furnace device process and gas pipe network system and perform data preprocessing; B. Intelligent prediction model of sulfur dioxide emission concentration uses process parameters to predict the sulfur dioxide emission concentration of the output heating furnace; C. Taking the predicted value of sulfur dioxide emission concentration of heating furnace as the judgment standard, construct the gas production device-gas pipeline network-gas consumption device gas pipeline network optimization scheduling model; D. Output the optimized scheduling strategy and scheduling plan for the gas system of each production unit within the preset time period in the future.

7. The method of the intelligent system for gas desulfurization based on scheme optimization according to claim 6 is characterized in that: The data preprocessing described in step A includes: real-time acquisition of all point data, deleting data during downtime, and performing process parameter preprocessing; The threshold value of the process parameters after cleaning in the previous step is calculated. The process parameters included in the model are uniformly calculated after being screened in the above steps, and a value range of each parameter is obtained. The value range is adjusted in combination with the properties of each process parameter and their actual rated working conditions. The value range of each process parameter is determined. The process parameters used for model construction are the process parameters after the value range is divided by the process parameter screening module; The data preprocessing described in step A includes: performing key parameter analysis on the relevant process parameters processed in the previous step, and incorporating the mined key parameters into the model.

8. The method of the intelligent system for gas desulfurization based on scheme optimization according to claim 6 is characterized in that: The gas network optimization scheduling model described in step C determines the gas production and consumption devices and the gas transportation network according to the scheduling optimization process of the gas network system, determines the desulfurization treatment facilities and processes used by the gas consumption devices, and establishes a production-transportation-consumption and waste gas treatment facility gas network optimization scheduling model; The reward and penalty function of the scheme evaluation under the current dispatching scheme is adopted, and the predicted value of sulfur dioxide emission concentration is used as the evaluation criterion; According to the received sulfur dioxide emission concentration forecast value, determine the number of heating furnaces exceeding the standard in the current state; According to the preset plan, adjust the gas dispatching plan of the heating furnace whose predicted sulfur dioxide emission concentration exceeds the standard; Recalculate the data after the adjustment of the dispatching plan and predict the output sulfur dioxide emission concentration. If the predicted sulfur dioxide emission concentration is less than the limit, the gas dispatching plan will be transmitted to the client for display; If the recalculated predicted value of sulfur dioxide emission concentration is still higher than the limit, the gas network optimization scheduling model will be returned to re-scheduling the plan until the predicted value of sulfur dioxide emission concentration is lower than the limit; It is not activated when the predicted sulfur dioxide emission concentration is lower than the limit.

9. The method of the intelligent system for gas desulfurization based on scheme optimization according to claim 6, characterized in that: The step D described in the output of the optimized scheduling strategy and scheduling plan of the gas system of each production unit within the preset time period in the future is specifically for the state that the predicted value of sulfur dioxide emission concentration exceeds the standard under the current gas scheduling plan, and the gas pipeline optimization scheduling model adjusts the gas scheduling plan. If the predicted value of sulfur dioxide emission concentration is lower than the limit value under this plan, the adjusted scheduling plan will be transmitted to the client port for display, so that the staff can perform optimized scheduling operations.

Images