Environment-friendly power generation waste incineration intelligent production scheduling optimization method

Abstract

An intelligent production scheduling optimization method for environment-friendly power generation waste incineration comprises the steps that S1, according to boiler operation and related data of waste, an objective function and constraint conditions which contribute to maximum total gross benefit (TCM) are established; s2, judging whether the next state of the boiler meets a constraint condition or not according to the related data of the boiler; s3, generating a start-stop scheduling sequence of each boiler; s4, calculating a start-stop sequence and an inventory ratio with maximized revenue; the algorithm is combined with the continuous start-stop period of each boiler, the daily start-stop state is judged and selected, and the loss caused by the fact that the boiler is turned from shutdown to operation is reduced. The boiler efficiency is associated with the inventory ratio, so that the boiler efficiency during incineration is improved, and the heat value of garbage combustion is increased. The data volume for arranging the start-stop state of equipment in a long time period is huge, so that the algorithm greatly reduces unreasonable data by optimizing the generation of a start-stop matrix, and an efficient and economical environment-friendly power plant production scheduling mode is realized by calculating and solving the total contribution gross benefit (TCM) through multiple iterations.

Description

Technical Field

[0001] The present invention relates to the field of waste treatment, and in particular to an intelligent production scheduling optimization method for environmentally friendly power generation waste incineration. Background Art

[0002] Waste incineration is an important method for efficiently, reducing and harmlessly treating domestic waste. While domestic waste is being disposed of in an environmentally friendly manner, the heat generated can also be used to generate electricity. It has now become the mainstream method for treating domestic waste in cities and towns in my country. The "Notice on the Installation of Automatic Pollutant Emission Monitoring Equipment and Networking in Domestic Waste Incineration Plants" issued by the Ministry of Ecology and Environment requires that the automatic monitoring data of pollutant emissions from domestic waste incineration power plants with scale in China be connected to the national environmental protection testing platform, and in accordance with the "Fixed Pollution Source Flue Gas (SO2, NO x The "Technical Specification for Continuous Monitoring of Emissions of Municipal Solid Waste (MSW) and Particulate Matter" marks the operating status of the municipal solid waste incinerator corresponding to the networked data, including the start-up and shutdown status of the municipal solid waste incinerator. The current "Standard for Control of Pollutants from Municipal Solid Waste Incineration" stipulates that the start-up process of the municipal solid waste incinerator starts from the input of domestic waste until the rated incineration volume is reached; the shutdown process of the municipal solid waste incinerator starts from the cessation of the input of domestic waste until the remaining domestic waste is completely burned. The start-up and shutdown status of the municipal solid waste incineration power plant is mainly assisted by the online monitoring of oxygen content, flow rate, flue gas temperature and other data at the end of the municipal solid waste incineration flue gas purification facility. Especially when the boiler needs periodic maintenance, most of the time, how to arrange the start and stop status of each boiler is judged based on the actual operating experience of the operating personnel of the municipal waste incineration power plant. However, there are differences in the personal experience and cognitive level of the operating personnel of each municipal waste incineration plant, and there are deviations in the judgment of the start and stop process. There is a risk of environmental liability in manual judgment, which makes it difficult to plan the start and stop sequence of the boiler during maintenance. In the existing technology, the problems caused by frequent start and stop of boilers during maintenance directly lead to low overall boiler efficiency and reduce the efficiency of all links in production. Therefore, it is very necessary to propose an intelligent production scheduling optimization method for environmentally friendly power generation waste incineration. Summary of the invention

[0003] The purpose of the present invention is to provide an intelligent production scheduling optimization method for environmentally friendly power generation waste incineration, which can reduce the losses caused by frequent start and stop, further improve the efficiency of boiler incineration waste treatment, and solve the existing technical defects and unattainable technical requirements.

[0004] To achieve the above object, the present invention provides the following technical solution: an environmentally friendly power generation waste incineration intelligent production scheduling optimization method, comprising the following steps:

[0005] S1. Based on the relevant data of boiler operation and garbage storage, establish the objective function and constraints for maximizing the total contribution gross profit (TCM)

[0006] 1. Obtain the operating parameters of each boiler, the historical amount of waste entering the site, and the most recent continuous start and stop time;

[0007] In this application, the boiler operating parameters that need to be obtained include: furnace calorific value, plant power consumption rate, standard boiler efficiency, steam turbine heat consumption, rated incineration capacity, boiler startup cost, ton supply, boiler efficiency coefficient and the actual incineration fuel amount of each boiler.

[0008] 1.2. Since the moisture content of garbage will greatly affect the calorific value of garbage combustion, the standard boiler efficiency is corrected through the large amount of data in step 1.1 to establish the relationship between the corrected boiler efficiency and the inventory ratio;

[0009] 1.3. Establish the objective function and constraints for maximizing the total contribution margin (TCM);

[0010] In this application, the total gross profit contribution (TCM) of the power plant is used as the objective function, the boiler start and stop sequence is used as the decision variable, and the boiler start and stop time and operation rules are used as constraints to establish an intelligent scheduling algorithm.

[0011] S2. Based on the relevant data of the boiler, determine whether the next state of the boiler meets the constraints

[0012] 2.1. Obtain the current start and stop status information of a boiler, and obtain the continuous start and stop time of the previous boiler based on this status information;

[0013] 2.2. According to the continuous start-stop duration in 2.1, set the next start-stop flag;

[0014] 2.3. Determine whether the generation of the next state satisfies the constraint condition of maximizing the total contribution gross profit (TCM) in step 1.3;

[0015] S3. Generate the start and stop scheduling sequence for each boiler (decision variables)

[0016] 3.1. Generate the start and stop scheduling sequence of each boiler according to the constraints and rules;

[0017] It should be noted in this application that the rules are:

[0018] (1) Each boiler has a minimum start-up time, a maximum start-up time, a minimum shutdown time, and a maximum shutdown time;

[0019] (2) Each boiler has continuous operation cycles and continuous shutdown cycles, and the normal distribution function is used to randomly distribute the expectations;

[0020] (3) In case of force majeure, such as power grid or boiler maintenance, the system must be shut down and the start-stop sequence must be set to 0;

[0021] (4) Initializing the start / stop flag of each boiler and the most recent continuous start / stop cycle, selecting the longest operation cycle and the longest shutdown cycle of the boiler based on a random selection method and comparing them with the most recent continuous start / stop cycle to determine the next start / stop flag;

[0022] (5) When the boiler is out of service for a long enough time, in order to prevent equipment damage and accurately adjust inventory, it is necessary to carefully consider whether to start the boiler;

[0023] (6) In order to ensure that at least one boiler is in operation and to exclude abnormal conditions where all boilers are shut down, the boilers must be switched to a shutdown state through conditional probability when they have been running for a long enough time.

[0024] 3.2. Introduce load reduction operation and strengthen the rules of boiler operation cycle;

[0025] 3.3. Combine the historical daily garbage inflow in step 1.1 and the amount of fuel burned by each boiler in the boiler operating parameters to calculate the impact and change of each boiler on the fuel inventory;

[0026] S4. Calculate the start-stop sequence and inventory ratio to maximize revenue

[0027] 4.1. Substitute the current operating parameters of all boilers into the objective function in step 1.3;

[0028] 4.2. Through multiple iterative optimizations, the objective function value is maximized and the corresponding start-stop sequence and inventory ratio are output.

[0029] Preferably, in step 1.2, the content of correcting the standard boiler efficiency is: correcting the standard boiler efficiency by a piecewise function to obtain a corrected boiler efficiency E c for:

[0030]

[0031] Among them, R best represents the optimal storage capacity ratio, R t It represents the inventory ratio on day t, and the calculation formula is: S t is the actual inventory on day t, S full Full stock quantity.

[0032] Preferably, in step 1.3, the specific contents are:

[0033] 1.3.1. The boiler start-stop matrix is ​​used as the decision variable X = [x1, x2, ..., x 365 ], where x t =[x 1，t , …, xi，t ] T , respectively represent the start and stop status of each boiler on the tth day, and the value is 0 or 1, 0 means shutdown, and 1 means operation;

[0034] 1.3.2. Establish the objective function max P to maximize the total contribution profit (TCM) inc (X), represents the total contribution margin (TCM) of the whole factory;

[0035] 1.3.3. Establish three constraints for maximizing the total contribution margin (TCM).

[0036] Preferably, the three TCM maximization constraints of step 1.3.3 include:

[0037] 1.3.3.1, 0<S t <S full , indicating that the changes in the warehouse caused by the burning and the amount of waste entering the factory make the actual inventory neither zero nor overflowing;

[0038] 1.3.3.2 and 0<x 1，t +…+x i，t ≤C, where C is the maximum number of boilers operating at the same time (according to the power plant contract), excluding the abnormal operating condition where all boilers are shut down;

[0039] 1.3.3.3、T on ~N(180,15 2 ), T off ~N(16,2.33 2 ), where the random variable T on represents the continuous operation cycle, the random variable T off represents the outage period, and the two are defined as discrete normal distribution random variables with 180 and 16 as the center points, and their normal distribution is recorded as N(μ, σ 2 ), where μ is the mean, σ 2 is the variance.

[0040] Preferably, in said S2, the specific contents are:

[0041] 2.1、Get the current operating status of a boiler, whether it is shut down or running. If it is currently shut down, get the number of consecutive days of shutdown from the current one; if it is currently running, get the number of consecutive days of running from the current one;

[0042] 2.2. According to the continuous start-stop duration in 2.1, set the next start-stop flag bit that is the same as the current state;

[0043] 2.3. Determine whether the generation of the next state satisfies the constraint condition of maximizing the total contribution gross profit (TCM) in step 1.3.3.3. If the constraint condition is met, the state remains unchanged; otherwise, the state is selected through the conditional probability function value.

[0044] Preferably, in step 2.3, when the generation of the next state does not satisfy the constraint condition, the specific content of state selection through the conditional probability function value is:

[0045] 2.3.1. When the boiler is running continuously, the current start-stop sequence is 1. If the constraint condition is met, the next start-stop sequence is set to 1, otherwise it is set to 0 with a certain probability. Y represents the value of the next state 0 or 1, and the probability function of Y is expressed as:

[0046]

[0047] Among them, T recent Indicates the most recent operation cycle, T start Indicates a continuous operating cycle;

[0048] 2.3.2. When the boiler is shut down continuously, the most recent start-stop sequence is 0. If the constraint condition is met, the most recent start-stop sequence is set to 0, otherwise it is set to 1 with a certain probability. The probability function of Y is expressed as:

[0049]

[0050] Among them, T recent Represents the most recent outage period, T stop Indicates consecutive outage periods.

[0051] Preferably, the specific content of step 3.2 is: the incineration amount of garbage is operated at a reduced load according to the calculated frequency, and a single calculation becomes five load reduction operations, wherein the load reduction operation rules are:

[0052] L i，k =L i ×(1-0.025×(k mod 5))

[0053] L k =[L 1，k ,…,L i，k ]

[0054] L i represents the fixed combustion load of the i-th boiler, L i,k It represents the amount of waste incineration of the i-th boiler at the k-th iteration. The value of k mod 5 is 0, 1, 2, 3, 4.

[0055] Preferably, in step 3.3, the inventory change is calculated as follows:

[0056] ΔS t =I t -∑L i，k

[0057] Where I t represents the amount of waste entering the factory on day t, △S t Indicates the change in the amount of garbage in the warehouse.

[0058] S t =S t-1 +ΔS t

[0059] 0 t full

[0060] In the formula, S t represents the amount of waste in the warehouse on day t, S full Indicates full inventory. When an abnormal situation occurs such as overflow or negative inventory, the start-stop sequence is regenerated.

[0061] Preferably, in step 4.1, the specific contents are:

[0062] 4.1.1. Calculate the daily inventory ratio and actual boiler efficiency E based on the daily inventory and storage capacity. The actual boiler efficiency E is calculated from the standard boiler efficiency E. s and corrected boiler efficiency E c Addition;

[0063] 4.1.2. Calculation of electricity revenue per ton of garbage P ce , tons of garbage subsidy B subs and future earnings estimate P pred ;

[0064] 4.1.3 Calculation of variable cost per ton of garbage B cost and total start-stop loss P cost ;

[0065] 4.1.4 Calculation of marginal contribution P per ton of garbage w , the total marginal contribution P and the total contribution margin (TCM) P inc .

[0066] Preferably, in step 4.1.2, P ce = α ×E a ×P e , represents the revenue from electricity generation through the heat generated by burning tons of garbage, where α Indicates tons of supply and boiler efficiency coefficient, E a Indicates the actual boiler efficiency, P e Indicates the electricity cost per kilowatt-hour; ​​

[0067] In step 4.1.4:

[0068] P w =α×E a ×P e +B subs -B cost

[0069] P=∑P w × t ×L k +P pred

[0070] P inc =P-∑P cost

[0071] P inc =∑(α×E a ×P e +B subs -B cost )×x t ×L k -∑P cost +P pred .

[0072] Compared with the prior art, the present invention has the following beneficial effects:

[0073] 1. The algorithm in this application combines the continuous start-stop cycle of each boiler to judge and select the daily start-stop status, and reduce the losses caused by the boiler shutdown to operation. By associating the boiler efficiency with the inventory ratio, the furnace efficiency during incineration is improved, and the calorific value of garbage combustion is increased. The amount of data for arranging the start-stop status of equipment over a long period of time is huge, so the algorithm greatly reduces unreasonable data by optimizing the generation of the start-stop matrix. After multiple iterations, the total contribution gross profit (TCM) is calculated and the solution to maximize it is sought, realizing an efficient and economical thermal power plant scheduling mode.

[0074] 2. This application can help the algorithm establish the start and stop status of the boiler more quickly through digital coding, facilitate the calculation of the start and stop flag, and determine the start and stop status requirements of the boiler at different time nodes.

[0075] 3. This application takes into account the start and stop requirements of the boiler during the initialization phase, and controls the generated start and stop sequence so that long-term shutdown and long-term operation, as well as frequent start and stop conditions, are avoided, thereby avoiding operating losses caused by this situation, directly reducing the start and stop sequences that do not meet the requirements in subsequent calculations, and reducing the workload of the optimization algorithm.

[0076] 4. This application uses a large amount of data experience from ten power plants to summarize the correction formula for boiler efficiency, comprehensively considers the impact of boiler start-up and shutdown on waste incineration efficiency, and more accurately calculates the marginal contribution of boiler incineration.

[0077] 5. This application controls the inventory ratio to achieve the optimal range of waste incineration calorific value and increase the waste incineration power generation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 Arrange the logic schematic diagram in the present invention;

[0079] Figure 2 This is a data chart of daily garbage entering the factory in Example 1 of the present invention;

[0080] Figure 3 It is a piecewise function diagram of the corrected furnace efficiency in Example 1 of the present invention;

[0081] Figure 4 It is a frequency distribution diagram of the continuous operation cycle in Example 1 of the present invention;

[0082] Figure 5 is a frequency distribution diagram of the outage period in Example 1 of the present invention;

[0083] Figure 6 The continuous operation cycle frequency distribution diagram of the first, second and third boilers in Example 1 of the present invention;

[0084] Figure 7 is a frequency distribution diagram of the shutdown period of the first boiler in Example 1 of the present invention;

[0085] Figure 8 is a frequency distribution diagram of the shutdown period of the second boiler in Example 1 of the present invention;

[0086] Fig. 9 is a frequency distribution diagram of the shutdown period of the third boiler in Example 1 of the present invention;

[0087] Fig.10 The start-stop sequence table of the first boiler when maximizing the revenue in Example 1 of the present invention;

[0088] Fig.11 The start-stop sequence table of the second boiler when maximizing the revenue in Example 1 of the present invention;

[0089] Fig.12 The start-stop sequence table of the third boiler when maximizing the revenue in Example 1 of the present invention;

[0090] Fig.13 It is the daily inventory ratio data diagram in Example 1 of the present invention; DETAILED DESCRIPTION

[0091] The following will be combined with the attached embodiment of the present invention Figure 1-13 , the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0092] See also Figure 1-13 , embodiments of the present invention:

[0093] Embodiment 1:

[0094] like Figure 1 As shown: An environmentally friendly power generation waste incineration intelligent production scheduling optimization method includes the following steps:

[0095] S1. Based on the boiler operation and garbage related data, establish the objective function and constraints for maximizing the total contribution gross profit (TCM)

[0096] 1.1. Obtain the operating parameters of each boiler, the historical daily waste volume entering the site, and the most recent continuous start and stop time; e.g. Figure 2

[0097] Here, the data obtained in this embodiment is a large amount of data from ten power plants, which can be understood as that this embodiment uses a large amount of data from ten power plants to empirically summarize the correction formula for boiler efficiency, comprehensively considers the impact of boiler start-up and shutdown on waste incineration efficiency, and more accurately calculates the marginal contribution of boiler incineration.

[0098] 1.2. Correct the standard boiler efficiency using the large amount of data in step 1.1, and establish the relationship between the corrected boiler efficiency and the inventory ratio;

[0099] The content of the correction of the standard boiler efficiency is as follows: the standard boiler efficiency is corrected by a piecewise function to obtain the corrected boiler efficiency E c for:

[0100]

[0101] Among them, R best represents the optimal storage capacity ratio, R t It represents the inventory ratio on day t, and the calculation formula is: S t is the actual inventory on day t, S full Full stock quantity.

[0102] In this embodiment, the inventory ratio mainly affects the water content of the garbage, and thus affects the calorific value generated when the garbage is incinerated. During the garbage stacking process, the interstitial water contained in the garbage is pressed out under the gravity of the upper garbage, and finally forms garbage leachate and is discharged from the warehouse. The reduction of the water content of the garbage is conducive to the improvement of the calorific value, that is, higher boiler efficiency. Through a large number of tests and operation management experience of a certain power plant, the optimal inventory ratio was obtained to be 0.62; so the furnace efficiency curve function: optimal storage capacity = 0.62, as shown in the following figure. Figure 3 As shown. In addition, when considering zero inventory and burst inventory, the water content in the garbage is the highest and the boiler efficiency is the lowest. The influence of inventory ratio on boiler efficiency is explored to construct a function of the relationship between boiler efficiency and inventory ratio. This function can calculate the boiler efficiency of each boiler under the daily inventory ratio. Taking into account the non-linear relationship of the function, a function curve that is closer to the actual combustion situation of the thermal power plant boiler is adopted, which in turn affects the final optimization goal and guides the production practice of the project.

[0103] 1.3. Establish the objective function and constraints for maximizing the total contribution margin (TCM);

[0104] The specific contents are:

[0105] 1.3.1. The boiler start-stop matrix is ​​used as the decision variable X = [x1, x2, ..., x 365 ], where x t =[x 1，t , …, x i，t ] T , respectively represent the start and stop status of each boiler on the tth day, and the value is 0 or 1, 0 means shutdown, and 1 means operation;

[0106] 1.3.2. Establish the objective function max P to maximize the total contribution profit (TCM) inc (X), represents the total contribution margin (TCM) of the whole factory;

[0107] 1.3.3. Establish three constraints for maximizing the total contribution margin (TCM).

[0108] The three TCM maximization constraints of step 1.3.3 are:

[0109] 1.3.3.1, 0<S t <S full , indicating that the changes in the warehouse caused by the burning and the amount of waste entering the factory make the actual inventory neither zero nor overflowing;

[0110] 1.3.3.2 and 0<x 1，t +…+x i，t≤C, where C is the maximum number of boilers running at the same time (according to the power plant contract agreement), and excludes the abnormal operating condition of all boilers being shut down. In this embodiment, C is 3, that is, 0<x 1，t +…+x i，t ≤3;

[0111] 1.3.3.3、T on ~N(180,15 2 ), T off ~N(16,2.33 2 ), where the random variable T on represents the continuous operation cycle, the random variable T off represents the outage period, and the two are defined as discrete normal distribution random variables with 180 and 16 as the center points, and their normal distribution is recorded as N(μ, σ 2 ), where μ is the mean, σ 2 is the variance.

[0112] Specifically, according to the engineering experience and historical data of boiler operation management, the continuous operation cycle is generally 135-225 days, and the shutdown maintenance cycle is generally 9-23 days. Most of the continuous operation time and maintenance time will be concentrated near a certain value (here the shutdown cycle is 16 days and the continuous operation cycle is 180 days). Using the normal distribution idea of ​​random variables, the discretized normal distribution is used to describe the variables, that is, the probability of each shutdown maintenance of about 16 days is the highest, and the probability of shutdown close to 9 days or 23 days is the lowest. The discretized normal distribution conforms to the actual working conditions, the simulation calculation is closer to the actual situation, and the optimization results are more credible and engineering valuable.

[0113] That is: T on ~N(180,15 2 ), T off ~N(16,2.33 2 ). The distribution diagram of continuous operation cycle and shutdown cycle is shown in Figure 4 and 5 As shown. Due to the differences between boilers, the actual shutdown period of the boiler is a subset of the defined probability distribution interval, thereby excluding impossible values. According to the actual operating conditions, each boiler has the longest and shortest shutdown period and the longest and shortest operation period. The probability distribution intervals of the shutdown period of boilers 1, 2, and 3 are [9, 22], [12, 22], and [14, 22], respectively. The probability distribution intervals of the continuous operation period of boilers 1, 2, and 3 are all [120, 210]. By intercepting the continuous operation period T on and outage period T off The normal distribution interval generates subintervals that can meet the actual situation.

[0114] Specifically, for boiler No. 1: Ton1 ~N(180,15 2 ), T on1 ∈[120,210], T off1 ~N(16,2.33 2 ), T off1 ∈[9,22]. For boiler No. 2: T on2 ~N(180,15 2 ), T on2 ∈[120,210], T off2 ~N(16,2.33 2 ), T off2 ∈[12,22]. For boiler No. 3: T on3 ~N(180,15 2 ), T on3 ∈[120,210], T off3 ~N(16,2.33 2 ), T off3 ∈[14,22]. Figure 6 is the distribution diagram of the discrete random variable values ​​of each boiler. The continuous operation cycle values ​​of boilers 1, 2, and 3 are in the range of [120, 210], and the frequency close to 180 is relatively large; Figure 7-9 As shown: the shutdown period values ​​of boilers 1, 2 and 3 fall within the ranges of [9, 22], [12, 22] and [14, 22] respectively, and the frequency close to 16 is relatively large.

[0115] S2. Based on the relevant data of the boiler, determine whether the next state of the boiler meets the constraints

[0116] 2.1、Get the current operating status of a boiler, whether it is shut down or running. If it is currently shut down, get the number of consecutive days of shutdown from the current one; if it is currently running, get the number of consecutive days of running from the current one;

[0117] 2.2. According to the continuous start-stop duration in 2.1, set the next start-stop flag bit that is the same as the current state:

[0118] 2.3. Determine whether the generation of the next state satisfies the constraint condition of maximizing the total contribution gross profit (TCM) in step 1.3.3.3. If the constraint condition is met, the state remains unchanged; otherwise, the state is selected through the conditional probability function value.

[0119] In step 2.3, when the generation of the next state does not meet the constraint conditions, the specific content of state selection through the conditional probability function value is:

[0120] 2.3.1. When the boiler is running continuously, the current start-stop sequence is 1. If the constraint condition is met, the next start-stop sequence is set to 1, otherwise it is set to 0 with a certain probability. Y represents the value of the next state 0 or 1, and the probability function of Y is expressed as:

[0121]

[0122] Among them, T recent Indicates the most recent operation cycle, T start Indicates a continuous operating cycle;

[0123] 2.3.2. When the boiler is shut down continuously, the most recent start-stop sequence is 0. If the constraint condition is met, the most recent start-stop sequence is set to 0, otherwise it is set to 1 with a certain probability. The probability function of Y is expressed as:

[0124]

[0125] Among them, T recent Represents the most recent outage period, T stop Indicates consecutive outage periods.

[0126] S3, generate the start and stop scheduling sequence for each boiler (such as Figure 10-12 )

[0127] 3.1. Generate the start and stop scheduling sequence of each boiler according to the constraints and rules;

[0128] 3.2. Introduce load reduction operation and strengthen the rules of boiler operation cycle;

[0129] The specific content is: the incineration volume of garbage is operated at a reduced load according to the calculated frequency, and a single calculation becomes five load reduction operations, among which the load reduction operation rules are:

[0130] L i，k =L i ×(1-0.025×(k mod 5))

[0131] L k =[L 1，k ,…,L i，k ]

[0132] L i represents the fixed combustion load of the i-th boiler, L i ,k represents the amount of waste incineration of the ith boiler at the kth iteration, and the value of kmod 5 is 0, 1, 2, 3, 4.

[0133] 3.3. Combine the historical daily garbage inflow in step 1.1 and the amount of fuel burned by each boiler in the boiler operating parameters to calculate the impact and change of each boiler on the fuel inventory;

[0134] In step 3.3, the inventory change is calculated as follows:

[0135] ΔS t =I t -∑L i，k

[0136] Where I t represents the amount of waste entering the factory on day t, △S t Indicates the change in the amount of garbage in the warehouse.

[0137] S t =S t-1 +ΔS t

[0138] 0<S t <S full

[0139] In the formula, S t represents the amount of waste in the warehouse on day t, S full Indicates full inventory. When an abnormal situation occurs such as overflow or negative inventory, the start-stop sequence is regenerated.

[0140] In this embodiment, in order to make the generated start-stop sequence meet the conditions of neither overflowing the inventory nor being negative, the inventory changes of the day are analyzed according to the daily garbage incineration volume and the garbage entering the plant, the daily inventory ratio is calculated, and it is judged whether the start-stop sequence generated this time is valid. Otherwise, the start-stop sequence is regenerated. The start-stop sequence generation method draws on the thinking of power plant managers and actual operation conditions, reduces the solution space, and increases the search probability of feasible solutions, so that a feasible solution is almost always found.

[0141] like Fig.13 As shown: S4, calculate the start-stop sequence and inventory ratio that maximizes revenue

[0142] 4.1. Substitute the current operating parameters of all boilers into the objective function in step 1.3;

[0143] The specific contents are:

[0144] 4.1.1. Calculate the daily inventory ratio and actual boiler efficiency E based on the daily inventory and storage capacity. The actual boiler efficiency E is calculated from the standard boiler efficiency E. s and corrected boiler efficiency E c Add, that is, E = E s +E c ;

[0145] 4.1.2. Calculation of electricity revenue per ton of garbage P ce , tons of garbage subsidy B subs and future earnings estimate P pred ;

[0146] Specific: P ce = α ×E a ×P e , represents the income from electricity generation through the heat generated by burning tons of garbage, where α represents the boiler efficiency coefficient, E a Indicates the actual boiler efficiency, P e Indicates the electricity cost per kilowatt-hour;

[0147] 4.1.3 Calculation of variable cost per ton of garbage B cost and total start-stop loss P cost ;

[0148] 4.1.4 Calculation of marginal contribution P per ton of garbage w , the total marginal contribution P and the total contribution margin (TCM) P inc ;

[0149] P w =α×E a ×P e +B subs -B cost

[0150] P=∑P w × t ×L k +P pred

[0151] P inc =P-∑P cost

[0152] P inc =∑(α×E a ×P e +B subs -B cost )×x t ×L k -∑P cost +P pred .

[0153] 4.2. Through 100,000 iterations of optimization, the objective function value is maximized and the corresponding start-stop sequence and inventory ratio are output.

[0154] Finally, the boiler start and stop control is carried out according to the output start and stop sequence and inventory ratio.

[0155] In this embodiment, the total contribution gross profit (TCM) generated by starting and stopping the boiler is: 190.286 million yuan

[0156] The working principle of this application can be roughly understood as follows:

[0157] (1) Boiler status initialization

[0158] First, read the amount of fuel entering the three boiler power plants and set the basic data of each boiler, including the amount of fuel burned, boiler startup cost, and other related basic parameters. Then, use the normal distribution model to estimate the length of time each boiler will stop for overhaul and run for a long period of time. During the scheduling process of each day, combined with the start and stop status, evaluate whether the following key rules are met: shortest start time, longest start time, shortest shutdown time, and longest shutdown time.

[0159] (2) Random generation of start-stop sequences

[0160] By evaluating the start and stop status of the boiler and the length of the most recent continuous start and stop cycle on a daily basis, it is recursively extended to 365 days throughout the year. During this period, a start and stop matrix for each boiler is constructed. The elements in the matrix are 0 or 1, representing that the boiler is in a shutdown or start state on that day. Finally, a 0-1 matrix of 365 days in length is generated for the three boilers. Combined with the daily amount of fuel entering the factory and the fuel consumption when the boiler is started and stopped, the impact and changes of each boiler on the fuel inventory are calculated. Among them, the incineration amount of garbage is operated at a reduced load according to the frequency of calculation, and a single calculation becomes five load reduction operations. Assuming that the total amount of garbage is insufficient, the boiler processing operation is reduced to increase the boiler operation cycle. In the event of abnormal conditions such as burst storage and negative inventory, the start and stop sequence should be regenerated.

[0161] (3) Optimal solution for marginal contribution within a cycle

[0162] The boiler efficiency curve is corrected by the inventory ratio, and the new boiler efficiency, waste incineration volume, and boiler start-up and shutdown loss fee are substituted into the marginal contribution total calculation formula. Through the brute force simulation method, tens of thousands of calculations are repeated to gradually screen out the optimal solution for maximizing the annual profit under these conditions, and output the optimal start-up and shutdown sequence and optimal inventory ratio for each boiler throughout the year. Finally, a visualization tool is used to display the daily fuel inventory change trend under the optimal solution, and determine the maximum profit achieved under this solution. Comparative Example 1:

[0163] The difference between this comparative example and Example 1 is that this comparative example uses manual production scheduling to regulate the start and stop of the boiler.

[0164] The specific content is: the boiler is shut down only when it needs maintenance and is started up after the maintenance is completed. The start-stop loss of the furnace and the impact of inventory on boiler efficiency are not considered, that is, multiple screening is not performed to seek maximum profit.

[0165] The total contribution gross profit (TCM) generated by regulating the start and stop of boilers in this comparative ratio by adopting manual production scheduling is: 188.183 million yuan.

[0166] In summary, the total contribution gross profit (TCM) generated by Example 1 is: 190.286 million yuan, while the total contribution gross profit (TCM) generated by Comparative Example 1 is: 188.183 million yuan, that is, an additional profit of 2.103 million yuan. The intelligent production scheduling algorithm is used to control the start-stop state, so that the start-stop sequence generated in almost every iteration is screened by the algorithm. In engineering, the boiler efficiency is improved and the frequent start-stop losses of the boiler are reduced. In terms of algorithm, the problem of space and computational explosion due to brute force measurement is avoided. After 100,000 iterations of optimization, the maximum value of the total contribution gross profit (TCM) is more than 2 million yuan higher than the total contribution gross profit (TCM) value of a single iteration. It is obvious that under the same conditions, the effect of Example 1 is better than that of Comparative Example 1 in terms of efficiency, total contribution gross profit (TCM), convenience, accuracy, etc., and after actual testing, about 10,000 groups of high-quality sequences can be generated every 3-5 minutes, and it is convenient and quick to select the approximate optimal solution from them, replacing the manual and experience work of the production scheduler.

[0167] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic features of the present invention. Therefore, no matter from which point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the attached claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any figure mark in the claims should not be regarded as limiting the claims involved.

[0168] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation modes that those skilled in the art can understand.

Claims

1. An environmentally friendly power generation waste incineration intelligent production scheduling optimization method, characterized by comprising the following steps: S1. Based on the boiler operation and garbage related data, establish the objective function and constraints for maximizing the total contribution gross profit (TCM) 1.

1. Obtain the operating parameters of each boiler, the historical daily waste volume entering the site, and the most recent continuous start and stop time; 1.

2. Correct the standard boiler efficiency using the large amount of data in step 1.1, and establish the relationship between the corrected boiler efficiency and the inventory ratio; 1.

3. Establish the objective function and constraints for maximizing the total contribution margin (TCM); S2. Based on the relevant data of the boiler, determine whether the next state of the boiler meets the constraints 2.

1. Obtain the current start and stop status information of a boiler, and obtain the continuous start and stop time of the previous boiler based on this status information; 2.

2. According to the continuous start-stop duration in 2.1, set the next start-stop flag; 2.

3. Determine whether the generation of the next state satisfies the constraint condition of maximizing the total contribution gross profit (TCM) in step 1.3; S3. Generate the start and stop scheduling sequence for each boiler 3.

1. Generate the start and stop scheduling sequence of each boiler according to the constraints and rules; 3.

2. Introduce load reduction operation and strengthen the rules of boiler operation cycle; 3.

3. Combine the historical daily garbage inflow in step 1.1 and the amount of fuel burned by each boiler in the boiler operating parameters to calculate the impact and change of each boiler on the fuel inventory; S4. Calculate the start-stop sequence and inventory ratio to maximize revenue 4.

1. Substitute the current operating parameters of all boilers into the objective function in step 1.3; 4.

2. Through multiple iterative optimizations, the objective function value is maximized and the corresponding start-stop sequence and inventory ratio are output.

2. According to the environmental protection power generation waste incineration intelligent production scheduling optimization method of claim 1, it is characterized in that in the step 1.2, the content of correcting the standard boiler efficiency is: correcting the standard boiler efficiency by a piecewise function to obtain the corrected boiler efficiency E c for: Among them, R best represents the optimal storage capacity ratio, R t It represents the inventory ratio on day t, and the calculation formula is: S t is the actual inventory on day t, S full Full stock quantity.

3. According to the environmental protection power generation waste incineration intelligent production scheduling optimization method of claim 2, it is characterized in that in the step 1.3, the specific content is: 1.3.

1. The boiler start-stop matrix is ​​used as the decision variable X = [x1, x2, ..., x 365 ], where x t =[x 1，t , …x i，t ] T , respectively represent the start and stop status of each boiler on the tth day, and the value is 0 or 1, 0 means shutdown, and 1 means operation; 1.3.

2. Establish the objective function max P to maximize the total contribution profit (TCM) inc (X), represents the total contribution margin (TCM) of the whole factory; 1.3.

3. Establish three constraints for maximizing the total contribution margin (TCM).

4. According to claim 3, an environmentally friendly power generation waste incineration intelligent production scheduling optimization method is characterized in that: The three TCM maximization constraints of step 1.3.3 are: 1.3.3.1, 0<S t <S full , indicating that the changes in the warehouse caused by the burning and the amount of waste entering the factory make the actual inventory neither zero nor overflowing; 1.3.3.2 and 0<x 1，t +…+x i，t ≤C, where C is the maximum number of boilers running at the same time, excluding the abnormal operating condition where all boilers are shut down; 1.3.3.3、T on ~N(180,15 2 ), T off ~N(16,2.33 2 ), where the random variable T on represents the continuous operation cycle, the random variable T off represents the outage period, and the two are defined as discrete normal distributed random variables with 180 and 16 as the center points. Their normal distribution is recorded as in, is the mean, σ 2 is the variance.

5. The method for optimizing intelligent production scheduling of environmentally friendly power generation waste incineration according to claim 4 is characterized in that: In said S2, the specific contents are: 2.1、Get the current operating status of a boiler, whether it is shut down or running. If it is currently shut down, get the number of consecutive days of shutdown from the current time. If it is currently running, get the number of consecutive days of running from the current time. 2.

2. According to the continuous start-stop duration in 2.1, set the next start-stop flag bit that is the same as the current state: 2.

3. Determine whether the generation of the next state satisfies the constraint condition of maximizing the total contribution gross profit (TCM) in step 1.3.3.

3. If the constraint condition is met, the state remains unchanged; otherwise, the state is selected through the conditional probability function value.

6. The method for optimizing intelligent production scheduling of environmentally friendly power generation waste incineration according to claim 5 is characterized in that: In step 2.3, when the generation of the next state does not meet the constraint conditions, the specific content of state selection through the conditional probability function value is: 2.3.

1. When the boiler is running continuously, the current start-stop sequence is 1. If the constraint condition is met, the next start-stop sequence is set to 1, otherwise it is set to 0 with a certain probability. Y represents the value of the next state 0 or 1, and the probability function of Y is expressed as: Among them, T recent Indicates the most recent operation cycle, T start Indicates a continuous operating cycle; 2.3.

2. When the boiler is shut down continuously, the most recent start-stop sequence is 0. If the constraint condition is met, the most recent start-stop sequence is set to 0, otherwise it is set to 1 with a certain probability. The probability function of Y is expressed as: Among them, T recent Represents the most recent outage period, T stop Indicates consecutive outage periods.

7. An environmentally friendly power generation waste incineration intelligent production scheduling optimization method according to claim 1, 2, 3, 4, 5 or 6, characterized in that: The specific content of step 3.2 is: the incineration amount of garbage is operated at a reduced load according to the calculated frequency, and a single calculation becomes five load reduction operations, wherein the load reduction operation rules are: L i，k =L i ×(1-0.025×(k mod 5)) L k =[L 1，k …,L i，k ] L i represents the fixed combustion load of the i-th boiler, L i,k It represents the amount of waste incineration of the i-th boiler at the k-th iteration. The value of kmod 5 is 0, 1, 2, 3, 4.

8. The method for optimizing intelligent production scheduling of environmentally friendly power generation waste incineration according to claim 7 is characterized in that: In step 3.3, the inventory change is calculated as follows: ΔS t =I t -∑L i，k Where I t represents the amount of waste entering the factory on day t, △S t Indicates the change in the amount of garbage in the warehouse; S t =S t-1 +ΔS t 0<S t <S full In the formula, S t represents the amount of warehouse garbage on day t, S full Indicates full inventory. When an abnormal situation occurs such as overflow or negative inventory, the start-stop sequence is regenerated.

9. The method for optimizing intelligent production scheduling of environmentally friendly power generation waste incineration according to claim 8 is characterized in that: In step 4.1, the specific contents are: 4.1.

1. Calculate the daily inventory ratio and actual boiler efficiency E based on the daily inventory and storage capacity. The actual boiler efficiency E is calculated from the standard boiler efficiency E. s and corrected boiler efficiency E c Addition; 4.1.

2. Calculation of electricity revenue per ton of garbage P ce , tons of garbage subsidy B subs and future earnings estimate P pred ; 4.1.3 Calculation of variable cost per ton of garbage B cost and total start-stop loss P cost ; 4.1.4 Calculation of marginal contribution P per ton of garbage w , the total marginal contribution P and the total contribution margin (TCM) P inc .

10. The method for optimizing intelligent production scheduling of environmentally friendly power generation waste incineration according to claim 9 is characterized in that: In step 4.1.2, P ce =α×E a ×P e , represents the income from electricity generation through the heat generated by burning tons of garbage, where α represents the boiler efficiency coefficient, E a Indicates the actual boiler efficiency, P e Indicates the electricity cost per kilowatt-hour; In step 4.1.4: P w =α×E a ×P e +B subs -B cost P=∑P w ×x t ×L k +P pred P inc =P-∑P cost P inc =∑(α×E a ×P e +B subs -B cost )×x t ×L k -∑P cost +P pred 。

Images