An interruptible load control method and apparatus based on a belt conveyor system

Abstract

The application discloses a kind of interruptible load control method and device based on belt conveying system, the technical scheme provided in the present application forms multiple working scenes by collecting the conveying data of belt conveying system, and generating the initial inventory of stock bin and the starting time of interruptible load execution mode, and obtains the longest interruptible time by optimizing calculation for each scene, multiple scene results are used as learning samples, and each working state of belt conveying system is covered by sample, then, model learning is carried out based on learning sample, and the interruptible load prediction model for belt conveying system is established, to output the longest interruption end time corresponding to the working scene parameter using the interruptible load prediction model constructed, to execute the demand response control of belt conveying system according to the result output by interruptible load prediction model, so as to ensure that belt conveying system participates in interruptible demand response while ensuring production safety.

Description

Technical Field

[0001] This application relates to the field of interruptible load prediction technology, and in particular to an interruptible load control method and apparatus based on a belt conveyor system. Background Technology

[0002] A key feature of the new power system is its ability to integrate a high proportion of renewable energy sources such as wind and solar power. In maintaining the stable operation of the new power system, interruptible loads play an important role in demand response.

[0003] Interruptible loads are incentive-based demand response resources that can interrupt part or all of their load during peak load periods or system failures. Belt conveyors are highly efficient and cost-effective, making them one of the main pieces of equipment for transporting discrete materials over medium to long distances. Belt conveyors are typically equipped with buffer facilities such as silos and stockpiles, and their designed transport capacity has a large margin compared to production demand. They can interrupt operation to participate in demand response and are commonly used interruptible loads in production scenarios such as mining, power, and chemical industries.

[0004] However, current interruptible load forecasting is based on general interruptible load characteristics, focusing on load interruption costs and efficiency, with the goal of minimizing energy costs or maximizing benefits for industrial users. It does not take into account the differences in operating principles and usage scenarios of specific interruptible loads, resulting in technical problems of low accuracy in interruption demand response control. Summary of the Invention

[0005] This application provides an interruptible load control method and apparatus based on a belt conveyor system, which solves the technical problem of low accuracy in interruption demand response control in existing belt conveyor systems.

[0006] To address the aforementioned technical problems, the first aspect of this application provides an interruptible load control method based on a belt conveyor system, comprising:

[0007] Collect material conveying data from the belt conveyor system;

[0008] Based on the material conveying data, the equipment parameter range of the belt conveyor system, and the demand response rules for interruptible loads, the initial values ​​of the working scenario parameters of the belt conveyor system are generated.

[0009] Based on the initial values ​​of the working scenario parameters and the production constraints of the belt conveyor system, a day-ahead optimization scheduling model for the belt conveyor system is constructed.

[0010] By solving the day-ahead optimization scheduling model, the material storage capacity of the belt conveyor system when the day-ahead optimization scheduling model reaches the optimal solution is obtained. Based on the material storage capacity and the material conveying data of the belt conveyor system, the maximum interruption end time of the belt conveyor system in a single working scenario is calculated.

[0011] By aggregating the initial values ​​of work scenario parameters and the maximum interruption end time under multiple single work scenarios as model learning samples, an interruptible load prediction model for the belt conveyor system is constructed, so as to perform demand response scheduling for the interruptible load of the belt conveyor system through the interruptible load prediction model.

[0012] Preferably, generating initial values ​​for the operating scenario parameters of the belt conveyor system based on the material conveying data, the equipment parameter range of the belt conveyor system, and the demand response rules for interruptible loads specifically includes:

[0013] Based on the material conveying data, determine the initial values ​​of the transfer volume and the initial values ​​of the production transportation volume of the belt conveyor system;

[0014] Based on the equipment parameter range of the belt conveyor system, determine the inventory range of the first and second silos in the belt conveyor system, and randomly generate the initial inventory values ​​of the first and second silos based on the inventory range.

[0015] Based on the demand response rules for interruptible loads, the interruption period range of the belt conveyor system is determined, and based on the interruption period range, the initial value of the interruption start time of the belt conveyor system is randomly determined.

[0016] Preferably, the day-ahead optimization scheduling model is as follows:

[0017]

[0018] st S 1(L) ≤S 1(j) ≤S 1(H)

[0019] S 2(L) ≤S 2(j) ≤S 2(H)

[0020]

[0021]

[0022] T min ≤T j ≤T max

[0023] In the formula, and T represents respectively start The material levels in the first and second silos at any given time, wherein the first silo is located upstream of the second silo. and S is the balance coefficient. 1(j) S represents the real-time inventory of the first silo at sampling time j. 2(j) S represents the real-time inventory of the second silo at sampling time j. 1(L) S is the lower limit of the inventory in the first silo. 1(H) S represents the maximum storage capacity of the first silo. 2(L) S represents the lower limit of the inventory in the second silo. 2(H) T represents the upper limit of the second silo's storage capacity. j Let T be the transport volume of the belt conveyor system at sampling time j. min ,T max These are the lower and upper limits of the transport capacity of the belt conveyor system, respectively. Let v be the belt speed of conveyor i in the belt conveyor system at sampling time j. min(i) ,v max(i) These are the lower and upper limits of the belt speed of conveyor i in the belt conveyor system, respectively. Let q be the mass of material per unit length of conveyor belt on conveyor i at sampling time j. G_max(i) This represents the upper limit of the material mass per unit length of the conveyor belt of conveyor i.

[0024] Preferably, the step of calculating the maximum interruption end time of the belt conveyor system in a single working scenario based on the silo inventory and the material conveying data of the belt conveyor system specifically includes:

[0025] Based on the first silo inventory output by the optimized scheduling model, and combined with the relationship between the transfer volume of the first silo and the first silo inventory, the first maximum interruption end time is calculated. The first interruption end time critical point is the longest time that the first silo can receive transfer volume after the belt conveyor system is interrupted.

[0026] Based on the second silo inventory output by the previous day's optimized scheduling model, and combined with the relationship between the production and transportation volume of the second silo and the second silo inventory, the second maximum interruption end time is calculated. The second maximum interruption end time is the longest time that the inventory of the second silo can support the production process after the belt conveyor system is interrupted.

[0027] The smaller of the first maximum interruption end time and the second maximum interruption end time is taken as the maximum interruption end time of the belt conveyor system in a single working scenario.

[0028] Preferably, the interruptible load prediction model is as follows:

[0029]

[0030]

[0031] x={T IN j ,T F j ,S 1(0) ,S 2(0) ,T start}

[0032]

[0033] In the formula, T IN j Let T be the transfer volume data of the belt conveyor system at the j-th sampling time. F j S represents the production and transportation volume data of the belt conveyor system at sampling time j. 1(0) S represents the initial inventory of the first silo. 2(0) K represents the initial inventory of the second silo. th The number of model learning samples is Nd, the number of days is δ, the width of the Gaussian kernel function is δ, ω1 and ω2 are weight vectors, b1 and b2 are bias coefficients, x is the model learning sample in a single working scenario, and A is a set of model learning samples in multiple single working scenarios.

[0034] Meanwhile, a second aspect of this application provides an interruptible load control device based on a belt conveyor system, comprising:

[0035] The conveying data acquisition unit is used to collect material conveying data from the belt conveyor system.

[0036] The work scenario data generation unit is used to generate initial values ​​of the work scenario parameters of the belt conveyor system based on the material conveying data, the equipment parameter range of the belt conveyor system, and the demand response rules for interruptible loads.

[0037] The day-ahead optimization scheduling model construction unit is used to construct the day-ahead optimization scheduling model of the belt conveyor system based on the initial values ​​of the working scenario parameters and the production constraints of the belt conveyor system.

[0038] The day-ahead optimization scheduling model solving unit is used to obtain the silo inventory of the belt conveyor system when the day-ahead optimization scheduling model reaches the optimal solution through solving the day-ahead optimization scheduling model, and to calculate the maximum interruption end time of the belt conveyor system in a single working scenario based on the silo inventory and the material conveying data of the belt conveyor system.

[0039] The interruptible load prediction model construction unit is used to summarize the initial values ​​of working scenario parameters and the maximum interruption end time under multiple single working scenarios as model learning samples to construct an interruptible load prediction model for the belt conveyor system, so as to perform demand response scheduling for the interruptible load of the belt conveyor system through the interruptible load prediction model.

[0040] Preferably, the work scene data generation unit is specifically used for:

[0041] Based on the material conveying data, determine the initial values ​​of the transfer volume and the initial values ​​of the production transportation volume of the belt conveyor system;

[0042] Based on the equipment parameter range of the belt conveyor system, determine the inventory range of the first and second silos in the belt conveyor system, and randomly generate the initial inventory values ​​of the first and second silos based on the inventory range.

[0043] Based on the demand response rules for interruptible loads, the interruption period range of the belt conveyor system is determined, and based on the interruption period range, the initial value of the interruption start time of the belt conveyor system is randomly determined.

[0044] Preferably, the day-ahead optimization scheduling model is as follows:

[0045]

[0046] st S 1(L) ≤S 1(j) ≤S 1(H)

[0047] S 2(L) ≤S 2(j) ≤S 2(H)

[0048]

[0049]

[0050] T min ≤T j ≤T max

[0051] In the formula, and T represents respectivelystart The material levels in the first and second silos at any given time, wherein the first silo is located upstream of the second silo. and S is the balance coefficient. 1(j) S represents the real-time inventory of the first silo at sampling time j. 2(j) S represents the real-time inventory of the second silo at sampling time j. 1(L) S is the lower limit of the inventory in the first silo. 1(H) S represents the maximum storage capacity of the first silo. 2(L) S represents the lower limit of the inventory in the second silo. 2(H) T represents the upper limit of the second silo's storage capacity. j Let T be the transport volume of the belt conveyor system at sampling time j. min ,T max These are the lower and upper limits of the transport capacity of the belt conveyor system, respectively. Let v be the belt speed of conveyor i in the belt conveyor system at sampling time j. min(i) ,v max(i) These are the lower and upper limits of the belt speed of conveyor i in the belt conveyor system, respectively. Let q be the mass of material per unit length of conveyor belt on conveyor i at sampling time j. G_max(i) This represents the upper limit of the material mass per unit length of the conveyor belt of conveyor i.

[0052] Preferably, the day-ahead optimization scheduling model solving unit is specifically used for:

[0053] By solving the day-ahead optimization scheduling model, the silo inventory of the belt conveyor system when the day-ahead optimization scheduling model reaches the optimal solution is obtained;

[0054] Based on the first silo inventory output by the optimized scheduling model, and combined with the relationship between the transfer volume of the first silo and the first silo inventory, the first maximum interruption end time is calculated. The first interruption end time critical point is the longest time that the first silo can receive transfer volume after the belt conveyor system is interrupted.

[0055] Based on the second silo inventory output by the previous day's optimized scheduling model, and combined with the relationship between the production and transportation volume of the second silo and the second silo inventory, the second maximum interruption end time is calculated. The second maximum interruption end time is the longest time that the inventory of the second silo can support the production process after the belt conveyor system is interrupted.

[0056] The smaller of the first maximum interruption end time and the second maximum interruption end time is taken as the maximum interruption end time of the belt conveyor system in a single working scenario.

[0057] Preferably, the interruptible load prediction model is as follows:

[0058]

[0059]

[0060] x={T IN j ,T F j ,S 1(0) ,S 2(0) ,T start}

[0061]

[0062] In the formula, T IN j Let T be the transfer volume data of the belt conveyor system at the j-th sampling time. F j S represents the production and transportation volume data of the belt conveyor system at sampling time j. 1(0) S represents the initial inventory of the first silo. 2(0) K represents the initial inventory of the second silo. th The number of model learning samples is Nd, the number of days is δ, the width of the Gaussian kernel function is δ, ω1 and ω2 are weight vectors, b1 and b2 are bias coefficients, x is the model learning sample in a single working scenario, and A is a set of model learning samples in multiple single working scenarios.

[0063] As can be seen from the above technical solutions, this application has the following advantages:

[0064] The technical solution provided in this application forms multiple working scenarios by collecting conveying data from the belt conveyor system and generating initial inventory levels in the silos and the start time of interruptible load execution. For each scenario, optimization calculations are performed to obtain the maximum interruptible time. The results from multiple scenarios are used as learning samples, covering various working states of the belt conveyor system. Then, based on the learning samples, a model learning process is performed to establish an interruptible load prediction model for the belt conveyor system. Using the constructed interruptible load prediction model, the maximum interruption end time corresponding to the input working scenario parameters is output. Demand response control of the belt conveyor system is executed based on the output of the interruptible load prediction model, thereby ensuring production safety while the belt conveyor system participates in interruptible demand response. This solves the technical problem of low accuracy in existing belt conveyor system interruptible demand response control. Attached Figure Description

[0065] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0066] Figure 1 This is a structural diagram of a factory material belt conveyor system.

[0067] Figure 2 This is a flowchart illustrating an embodiment of an interruptible load control method based on a belt conveyor system provided in this application.

[0068] Figure 3 This is a flowchart illustrating step 102 of an interruptible load control method based on a belt conveyor system provided in this application.

[0069] Figure 4 This is a schematic diagram of the demand response principle for a production process based on interruptible loads using a belt conveyor system.

[0070] Figure 5 This is a flowchart illustrating step 104 of an interruptible load control method based on a belt conveyor system provided in this application.

[0071] Figure 6 This is a logical diagram of the interruptible load prediction model in an interruptible load control method based on a belt conveyor system provided in this application.

[0072] Figure 7 This is a schematic diagram of an embodiment of an interruptible load control device based on a belt conveyor system provided in this application. Detailed Implementation

[0073] This application provides an interruptible load control method and apparatus based on a belt conveyor system, which solves the technical problem of low accuracy in interruption demand response control in existing belt conveyor systems.

[0074] To make the inventive objectives, features, and advantages of this application more apparent and understandable, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0075] like Figure 1The diagram shows the structure of a factory belt conveyor system, a bulk material transportation system. Materials are transferred from a remote location to the first storage silo using conveyors, trucks, trains, etc., with a transfer capacity of T. IN Then through B1 to B N The conveyor line transports materials to the second silo, and finally, according to production needs, the bulk materials are sent to the factory for processing into semi-finished or finished products. The production transport volume is T. F B1~B N Conveyor lines can be quite long, reaching tens of kilometers; the conveyors are powerful, with a single conveyor capable of handling hundreds of kW, often employing dual or even multi-motor drives. Bulk material conveying systems are typically designed based on the factory's maximum production capacity, with significant margins. When the factory is not operating at full capacity, the daily working time of the bulk material conveying system is only a few hours or even a dozen hours. Furthermore, silos provide a substantial buffer for the bulk material conveying system. Figure 1 The typical factory material handling system shown is suitable for use as an interruptible load in demand response.

[0076] against Figure 1 The present application first provides a detailed description of an embodiment of an interruptible load control method based on a belt conveyor system, illustrating a typical factory material transport system.

[0077] Please see Figure 2 This embodiment provides an interruptible load control method based on a belt conveyor system, comprising:

[0078] Step 101: Collect material conveying data from the belt conveyor system.

[0079] It should be noted that the solution provided in this embodiment first collects material conveying data of the belt conveyor system based on its structural characteristics. More specifically, the material conveying data mentioned in this embodiment mainly includes: transfer volume T. IN and production and transportation volume T F Multi-day data curve, denoted as N days d These data essentially cover all operating conditions of the belt conveyor system, with a preferred sampling period of 15 minutes. The transfer volume data and production data are represented by matrices, denoted as follows: and Each row of the matrix represents one day's data, and each column contains 96 data points obtained at a 15-minute sampling period. It is understood that the 15-minute sampling period proposed in this embodiment is only an example. Users can adjust the value of the sampling period according to the actual application scenario requirements. Correspondingly, if the sampling period is different, the amount of data in each column of the matrix composed of material conveying data will also change accordingly.

[0080] Step 102: Generate initial values ​​for the working scenario parameters of the belt conveyor system based on the material conveying data, the equipment parameter range of the belt conveyor system, and the demand response rules for interruptible loads.

[0081] It should be noted that, based on the material conveying data obtained in step 101, combined with the equipment parameter range of the belt conveyor system and the demand response rules of interruptible loads, the initial values ​​of the working scenario parameters of the belt conveyor system are generated. Each set of initial values ​​of working scenario parameters represents a working scenario of the belt conveyor system.

[0082] In some embodiments, the working scenario parameters mentioned in step 102 include: initial value of transfer volume, initial value of production transportation volume, initial value of inventory in the first and second material bins, and initial value of interruption start time.

[0083] like Figure 3 As shown, the process of generating work scene parameters in step 102 may include the following steps:

[0084] Step 1021: Based on the material conveying data, determine the initial values ​​of the transfer volume and the initial values ​​of the production transportation volume of the belt conveyor system;

[0085] Step 1022: Based on the equipment parameter range of the belt conveyor system, determine the storage range of the first and second silos in the belt conveyor system, and randomly generate the initial storage values ​​of the first and second silos based on the storage range.

[0086] Step 1023: Determine the interruption period range of the belt conveyor system according to the demand response rules of interruptible loads, and randomly determine the initial value of the interruption start time of the belt conveyor system based on the interruption period range.

[0087] It should be noted that this embodiment selects the transshipment volume data for day i. and production and transportation volume Then, an initial inventory is randomly generated from the effective range of the first silo's inventory.

[0088] S 1(0) = rand()·(S 1(H) -S 1(L) )+S 1(L) (1)

[0089] In the formula, S 1(0) S 1(L) and S 1(H) These represent the initial inventory, lower limit, and upper limit of the first hopper, respectively. `rand()` is a random number generator that outputs a random number between 0 and 1. The initial inventory of the second hopper is generated in the same way.

[0090] S2(0) =rand()•(S 2(H) -S 2(L) )+S 2(L) (2)

[0091] In the formula, S 2(0) S 2(L) and S 2(H) These are the initial inventory, lower inventory limit, and upper inventory limit of the second silo, respectively. rand() is a random number generation function that outputs a random number between 0 and 1.

[0092] This embodiment also follows the interruptible load demand response rules stipulated in the "Implementation Rules of Market-Oriented Demand Response in Guangdong Province." When the dispatch center calls upon interruptible loads, it will notify power users in advance to prepare, providing a 2-hour preparation period. Each interruptible load response duration is no less than 2 hours. This embodiment predicts the interruptible time of the transmission system on a daily basis; therefore, the first and last 2 hours of the day need to be removed. Using a 15-minute sampling period, the possible range of the starting time for the interruptible load response of the transmission system within a day is T. start ∈[9,88]. The start point for the interruptible load response is randomly generated from the effective interval [9,88]. When a set of... S 1(0) S 2(0) And T start Then, a work scenario was determined.

[0093] Step 103: Based on the initial values ​​of the working scenario parameters and the production constraints of the belt conveyor system, construct a day-ahead optimization scheduling model for the belt conveyor system.

[0094] It should be noted that after determining a certain working scenario of the belt conveyor system in step 102, based on the initial values ​​of the working scenario parameters and the production constraints of the belt conveyor system, a day-ahead optimization scheduling model for the belt conveyor system that satisfies the production constraints is constructed. The construction of the day-ahead optimization scheduling model can be seen in the following example:

[0095] like Figure 4 As shown, the pre-day optimized scheduling is executed at 00:00 on the current day, achieving a load interruption time T while meeting production constraints. span Maximize. Energy calculation model for belt conveyors:

[0096]

[0097] In the formula, p(v,T) is the power of the belt conveyor (kW); v is the belt speed (m / s); and T is the conveyor capacity (t / h). The model coefficients a, b, c, d, and e are obtained from the design parameters, or through parameter identification or curve fitting. The daily energy consumption of the conveying system is expressed as:

[0098]

[0099] Where N is the number of conveyors, Δt is the sampling period (15 min), and T is the sampling time. j Let B1 be the transport volume at sampling time j, and B1 be the transport volume at sampling time j. N Conveyors connected in series form a transport line, therefore they have the same transport capacity. Let conveyor i have a speed at sampling time j. The inventory S in the first hopper... 1(j) This can be expressed as:

[0100]

[0101] The transfer volume at sampling time j on day K. The inventory S in the second silo. 2(j) This can be expressed as:

[0102]

[0103] Let T be the production and transportation volume at sampling time j on day K. Based on the analysis results in step 3.1, at time T... start T can be achieved by minimizing the inventory in the first hopper and maximizing the inventory in the second hopper at all times. span To maximize energy efficiency in the conveying system, the optimization also considers minimizing the total daily energy consumption. Therefore, the objective function of the optimization problem is expressed as the total daily energy consumption, T. start The combined inventory of Warehouse 1 and Warehouse 2 is as follows:

[0104]

[0105] In the formula, and T represents respectively start The amount of material stored in the first and second silos at any given time. and This is the balance coefficient. By minimizing the objective function, we can simultaneously minimize the total daily energy consumption and T. start Minimize the inventory in the first hopper and maximize the inventory in the second hopper at any given time, by changing... and The importance of the three optimization objectives can be adjusted. and Derived from equations (3) and (4), it is expressed as:

[0106]

[0107]

[0108] The current optimization problem also needs to meet some constraints: the material inventory in the first and second silos should always be within their safety limits, that is:

[0109] S 1(L) ≤S 1(j) ≤S 1(H) (10)

[0110] S 2(L) ≤S 2(j) ≤S 2(H) (11)

[0111] B1-B N The mass of material per unit length of conveyor belt must be less than or equal to its maximum value, that is:

[0112]

[0113] B1-B N The belt speed and conveying capacity of the conveyor should meet the corresponding physical limitations, the specific physical limitations are as follows:

[0114]

[0115] T min ≤T j ≤T max (j=1,…,96)(14)

[0116] Therefore, the day-ahead optimization scheduling model for the conveying system that satisfies production constraints can be summarized as follows:

[0117]

[0118] The optimization variables for this problem are conveyors B1-B. N The vector consisting of belt speed and transport volume is denoted as

[0119] Step 104: By solving the day-ahead optimization scheduling model, obtain the silo inventory of the belt conveyor system when the day-ahead optimization scheduling model reaches the optimal solution, and calculate the maximum interruption end time of the belt conveyor system in a single working scenario based on the silo inventory and the material conveying data of the belt conveyor system.

[0120] It should be noted that the day-ahead optimization problem of the conveyor system is solved based on the production constraints in step 103. By solving the optimization problem (15), the conveyor scheduling instruction can be obtained. This scheduling instruction can minimize the total daily energy consumption and T while satisfying the production constraints. start The goal is to minimize the inventory in the first silo and maximize the inventory in the second silo at any given time, and to obtain the inventory S of the first silo when the optimal solution is achieved. 1(optimal) and the inventory S in the second silo 2(optimal) The optimization problem (15) contains nonlinear terms and is solved using the fmincon function in Matlab.

[0121] like Figure 5 As shown, in some embodiments, step 104 of this embodiment may specifically include:

[0122] Step 1041: By solving the day-ahead optimization scheduling model, obtain the silo inventory of the belt conveyor system when the day-ahead optimization scheduling model reaches the optimal solution;

[0123] Step 1042: Based on the first silo inventory output by the optimized scheduling model, and combined with the relationship between the transfer volume and the first silo inventory, calculate the first maximum interruption end time. The critical point of the first interruption end time is the longest time that the first silo can receive transfer volume after the belt conveyor system is interrupted.

[0124] Step 1043: Based on the second silo inventory output by the optimized scheduling model, and combined with the relationship between the production and transportation volume of the second silo and the second silo inventory, calculate the second maximum interruption end time. The second maximum interruption end time is the longest time that the second silo inventory can support the production process after the belt conveyor system is interrupted.

[0125] Step 1044: Take the smaller value between the first maximum interruption end time and the second maximum interruption end time as the maximum interruption end time of the belt conveyor system in a single working scenario.

[0126] It should be noted that S was obtained by solving the optimization problem while satisfying production constraints. 1(optimal) and S 2(optimal) The following formula can be used to obtain the conveyor B1-B N The longest time T that the first silo can receive the transferred amount after the operation is interrupted. end(1) :

[0127]

[0128] Similarly, the conveyor B1-B can be obtained. N The maximum time (T) that the material inventory in the second silo can support the production process after an interruption of operation. end(2) :

[0129]

[0130] Conveyor B1-B N The longest time that interrupts operation is expressed as

[0131] T end(*) =min(T) end(1) ,T end(2) (18)

[0132] Step 105: Summarize the initial values ​​of work scenario parameters and the maximum interruption end time under multiple single work scenarios as model learning samples, and construct an interruptible load prediction model for the belt conveyor system. In order to use the interruptible load prediction model to perform demand response scheduling for the interruptible load of the belt conveyor system.

[0133] Finally, based on step 104, the longest interruption time of the belt conveyor load was obtained in a single scenario. The structure of the interruptible load prediction model for the production-constrained belt conveyor system established in this embodiment is as follows: Figure 6 As shown, it should be noted that and Let S be a vector. 1(0) S 2(0) And T start As scalars, they are combined to construct the input vector. The output is y = T end(*) , will {x s ,y} is added as a sample to the learning sample library. After obtaining a learning sample, step 102 can be repeated to randomly select S. 1(0) S 2(0) And T start Optimize the new scenario to obtain the longest load interruption time under the new scenario, and build new learning samples until the preset number of times K is reached. th Then, take the transshipment volume data and production transportation volume data from another day and perform optimization calculations to obtain new learning samples. After processing all N... d After obtaining the data for the day, K will be obtained. th ×N d A sample set is created from these samples to construct an interruptible load forecasting model.

[0134] Regarding the construction of the interruptible load forecasting model, this embodiment preferably uses a nonlinear weighted Lagrange ε-Twin Support Vector Regression Machine (WL-ε-TSVR) to establish the forecasting model. Define the matrix. For the input training samples, vector To output training samples, and to handle the nonlinearity of the model, this embodiment uses a Gaussian kernel function to map the training samples to a high-dimensional space for processing. The kernel function is expressed as:

[0135]

[0136] In the formula, δ is the width of the Gaussian kernel function. WL-ε-TSVR requires finding two decision hyperplanes, expressed as f1(x)=K(x) T A T )ω1+b1 and f2(x)=K(x T A T Let ω₁ and ω₂ be weight vectors, and b₁ and b₂ be bias coefficients; ω₁, ω₂, b₁, and b₂ are obtained by solving two QPPs. To improve algorithm performance, a weight matrix D is introduced into the original objective function. Using the Lagrangian function and combining it with the KKT conditions, the dual problem of the original objective function is obtained as follows:

[0137]

[0138]

[0139] In the formula, C1, C2, υ1, υ2, ε1, and ε2 are adjustment parameters; a and γ are Lagrange multipliers; I is the identity matrix; H = [K(A,A T e]; e = [l,…,l] is a vector of appropriate dimensions. By solving equations (20) and (21), we can obtain:

[0140] [ω1 b1] T =(H T H+υ1I) -1 H T (Ya) (22)

[0141] [ω2 b2] T =(H T H+υ2I) -1 H T (Y+γ) (23)

[0142] The final regression function is:

[0143]

[0144] Equation (24) is the interruptible load prediction model for a belt conveyor system under production constraints. The predicted daily bulk material transfer volume T is then used. IN j Production demand forecast T F j Initial inventory S in the first silo 1(0) Initial inventory S in the second silo 2(0) and the start time T of interruptible load demand response execution startConstruct the input vector x = {T IN j ,T F j ,S 1(0) ,S 2(0) ,T start Substituting this into model (24) allows us to predict the longest period T during which the load can be interrupted. end(*) = f(x).

[0145] The technical solution provided in this application forms multiple working scenarios by collecting conveying data from the belt conveyor system and generating initial inventory levels in the silos and the start times of interruptible load execution. For each scenario, optimization calculations are performed to obtain the maximum interruptible time. The results from multiple scenarios are used as learning samples, covering various working states of the belt conveyor system. Then, based on these learning samples, a model is learned, considering the structural characteristics and working methods of the belt conveyor system, to establish an interruptible load prediction model for the belt conveyor system. Using this model, the maximum interruption end time corresponding to the input working scenario parameters is output. Demand response control of the belt conveyor system is then executed based on the output of the interruptible load prediction model, ensuring production safety while participating in interruptible demand response and improving the accuracy of interruptible demand response control for the belt conveyor system.

[0146] The above is a detailed description of an embodiment of an interruptible load control method based on a belt conveyor system provided in this application. The following is a detailed description of an embodiment of an interruptible load control device based on a belt conveyor system provided in this application.

[0147] Please see Figure 7 This embodiment provides an interruptible load control device based on a belt conveyor system, comprising:

[0148] The conveying data acquisition unit 201 is used to collect material conveying data of the belt conveyor system.

[0149] The work scenario data generation unit 202 is used to generate initial values ​​of work scenario parameters for the belt conveyor system based on material conveying data, the equipment parameter range of the belt conveyor system, and the demand response rules for interruptible loads.

[0150] The day-ahead optimization scheduling model construction unit 203 is used to construct the day-ahead optimization scheduling model of the belt conveyor system based on the initial values ​​of the working scenario parameters and the production constraints of the belt conveyor system.

[0151] The day-ahead optimization scheduling model solving unit 204 is used to obtain the silo inventory of the belt conveyor system when the day-ahead optimization scheduling model reaches the optimal solution through solving the day-ahead optimization scheduling model, and to calculate the maximum interruption end time of the belt conveyor system in a single working scenario based on the silo inventory and the material conveying data of the belt conveyor system.

[0152] The interruptible load prediction model building unit 205 is used to summarize the initial values ​​of working scenario parameters and the maximum interruption end time under multiple single working scenarios as model learning samples to build an interruptible load prediction model for the belt conveyor system, so as to perform demand response scheduling for the interruptible load of the belt conveyor system through the interruptible load prediction model.

[0153] Furthermore, the work scenario data generation unit 202 is specifically used for:

[0154] Based on the material conveying data, determine the initial values ​​of the transfer volume and the initial values ​​of the production transportation volume of the belt conveyor system;

[0155] Based on the equipment parameter range of the belt conveyor system, determine the storage range of the first and second silos in the belt conveyor system, and randomly generate the initial storage values ​​of the first and second silos based on the storage range.

[0156] Based on the demand response rules for interruptible loads, the interruption period range of the belt conveyor system is determined, and based on the interruption period range, the initial value of the interruption start time of the belt conveyor system is randomly determined.

[0157] Furthermore, the current optimized scheduling model is as follows:

[0158]

[0159] st S 1(L) ≤S 1(j) ≤S 1(H)

[0160] S 2(L) ≤S 2(j) ≤S 2(H)

[0161]

[0162]

[0163] T min ≤T j ≤T max

[0164] In the formula, and T represents respectively startThe material levels in the first and second silos are measured at any given time, with the first silo located upstream of the second silo. and S is the balance coefficient. 1(j) S represents the real-time inventory of the first silo at sampling time j. 2(j) S represents the real-time inventory of the second silo at sampling time j. 1(L) S is the lower limit of the inventory in the first silo. 1(H) S represents the maximum storage capacity of the first silo. 2(L) S is the lower limit of the inventory in the second silo. 2(H) T is the upper limit of the second silo's storage capacity. j Let J be the transport volume of the belt conveyor system at sampling time j. Let J be the belt speed of conveyor i in the belt conveyor system at sampling time j. Let be the mass of material per unit length of conveyor belt on conveyor i at sampling time j.

[0165] Furthermore, the current optimized scheduling model solution unit 204 is specifically used for:

[0166] By solving the day-ahead optimization scheduling model, the silo inventory of the belt conveyor system when the day-ahead optimization scheduling model reaches the optimal solution is obtained;

[0167] Based on the first silo inventory output by the optimized scheduling model, and combined with the relationship between the first silo's transfer volume and its inventory, the first maximum interruption end time is calculated. The critical point of the first interruption end time is the longest time that the first silo can receive transfer volume after the belt conveyor system is interrupted.

[0168] Based on the second silo inventory output by the optimized scheduling model, and combined with the relationship between the production and transportation volume of the second silo and the second silo inventory, the second maximum interruption end time is calculated. The second maximum interruption end time is the longest time that the second silo inventory can support the production process after the belt conveyor system is interrupted.

[0169] The smaller of the first maximum interruption end time and the second maximum interruption end time is taken as the maximum interruption end time of the belt conveyor system in a single working scenario.

[0170] Furthermore, the interruptible load forecasting model is as follows:

[0171]

[0172]

[0173] x={T IN j ,T F j ,S 1(0) ,S2(0) ,T start}

[0174]

[0175] In the formula, T IN j Let T be the transfer volume data of the belt conveyor system at the j-th sampling time. F j S represents the production and transport volume data of the belt conveyor system at sampling time j. 1(0) S represents the initial inventory of the first silo. 2(0) K represents the initial inventory of the second silo. th denoted as the number of model learning samples, Nd as the number of days, δ as the width of the Gaussian kernel function, ω1 and ω2 as weight vectors, b1 and b2 as bias coefficients, x as the model learning sample in a single work scenario, and A as the set of model learning samples in multiple single work scenarios.

[0176] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described apparatus and unit can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0177] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0178] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the application described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0179] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0180] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0181] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0182] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0183] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for interruptible load control based on a belt conveyor system, characterized in that, include: Collect material conveying data from the belt conveyor system; Based on the material conveying data, the equipment parameter range of the belt conveyor system, and the demand response rules for interruptible loads, the initial values ​​of the working scenario parameters of the belt conveyor system are generated. Based on the initial values ​​of the working scenario parameters and the production constraints of the belt conveyor system, a day-ahead optimization scheduling model for the belt conveyor system is constructed. By solving the day-ahead optimization scheduling model, the material storage capacity of the belt conveyor system when the day-ahead optimization scheduling model reaches the optimal solution is obtained. Based on the material storage capacity and the material conveying data of the belt conveyor system, the maximum interruption end time of the belt conveyor system in a single working scenario is calculated. The initial values ​​of working scenario parameters and the maximum interruption end time under multiple single working scenarios are collected as model learning samples to construct an interruptible load prediction model for the belt conveyor system, so as to perform demand response scheduling for the interruptible load of the belt conveyor system through the interruptible load prediction model. Specifically, generating the initial values ​​of the working scenario parameters of the belt conveyor system based on the material conveying data, the equipment parameter range of the belt conveyor system, and the demand response rules for interruptible loads includes: Based on the material conveying data, determine the initial values ​​of the transfer volume and the initial values ​​of the production transportation volume of the belt conveyor system; Based on the equipment parameter range of the belt conveyor system, determine the inventory range of the first and second silos in the belt conveyor system, and randomly generate the initial inventory values ​​of the first and second silos based on the inventory range. Based on the demand response rules for interruptible loads, the interruption period range of the belt conveyor system is determined, and based on the interruption period range, the initial value of the interruption start time of the belt conveyor system is randomly determined. The interruptible load prediction model is specifically as follows: In the formula, This refers to the transfer volume data of the belt conveyor system at the j-th sampling time. This refers to the production and transportation volume data of the belt conveyor system at sampling time j. This represents the initial inventory level of the first silo. K represents the initial inventory of the second silo. th N represents the number of training samples for the model. d For the number of days, The width of the Gaussian kernel function. and It is a weight vector. and The bias coefficient, These are model learning samples for a single work scenario. T is a set of model learning samples for multiple single working scenarios. start This is the starting time for the daily interruptible load response of the transportation system.

2. The interruptible load control method based on a belt conveyor system according to claim 1, characterized in that, The day-ahead optimization scheduling model is specifically as follows: In the formula, and T represents respectively start The material levels in the first and second silos at any given time, wherein the first silo is located upstream of the second silo. and For balance coefficient, Let J be the real-time inventory of the first silo at sampling time j. This represents the real-time inventory of the second silo at sampling time j. This is the lower limit of the inventory in the first silo. This represents the maximum storage capacity of the first silo. This is the lower limit of the inventory in the second silo. This is the upper limit of the storage capacity of the second silo. Let j be the transport volume of the conveyor in the belt conveyor system at sampling time j. These represent the lower and upper limits of the conveyor's transport capacity in the belt conveyor system, respectively. Let $\frac{i}{j}$ be the belt speed of conveyor $i$ in the belt conveyor system at sampling time $j$. These are the lower and upper limits of the belt speed of conveyor i in the belt conveyor system, respectively. Let the mass of material per unit length of conveyor belt on conveyor i be the mass of material on conveyor i at sampling time j. This represents the upper limit of the material mass per unit length of the conveyor belt of conveyor i.

3. The interruptible load control method based on a belt conveyor system according to claim 2, characterized in that, The step of calculating the maximum interruption end time of the belt conveyor system under a single working scenario based on the silo inventory and the material conveying data of the belt conveyor system specifically includes: Based on the first silo inventory output by the optimized scheduling model, and combined with the relationship between the transfer volume of the first silo and the first silo inventory, the first maximum interruption end time is calculated. The critical point of the first maximum interruption end time is the longest time that the first silo can receive transfer volume after the belt conveyor system is interrupted. Based on the second silo inventory output by the previous day's optimized scheduling model, and combined with the relationship between the production and transportation volume of the second silo and the second silo inventory, the second maximum interruption end time is calculated. The second maximum interruption end time is the longest time that the inventory of the second silo can support the production process after the belt conveyor system is interrupted. The smaller of the first maximum interruption end time and the second maximum interruption end time is taken as the maximum interruption end time of the belt conveyor system in a single working scenario.

4. An interruptible load control device based on a belt conveyor system, characterized in that, include: The conveying data acquisition unit is used to collect material conveying data from the belt conveyor system. The work scenario data generation unit is used to generate initial values ​​of the work scenario parameters of the belt conveyor system based on the material conveying data, the equipment parameter range of the belt conveyor system, and the demand response rules for interruptible loads. The day-ahead optimization scheduling model construction unit is used to construct the day-ahead optimization scheduling model of the belt conveyor system based on the initial values ​​of the working scenario parameters and the production constraints of the belt conveyor system. The day-ahead optimization scheduling model solving unit is used to obtain the silo inventory of the belt conveyor system when the day-ahead optimization scheduling model reaches the optimal solution through solving the day-ahead optimization scheduling model, and to calculate the maximum interruption end time of the belt conveyor system in a single working scenario based on the silo inventory and the material conveying data of the belt conveyor system. The interruptible load prediction model construction unit is used to summarize the initial values ​​of the working scenario parameters and the maximum interruption end time under multiple single working scenarios as model learning samples to construct an interruptible load prediction model for the belt conveyor system, so as to perform demand response scheduling for the interruptible load of the belt conveyor system through the interruptible load prediction model. Specifically, the work scenario data generation unit is used for: Based on the material conveying data, determine the initial values ​​of the transfer volume and the initial values ​​of the production transportation volume of the belt conveyor system; Based on the equipment parameter range of the belt conveyor system, determine the inventory range of the first and second silos in the belt conveyor system, and randomly generate the initial inventory values ​​of the first and second silos based on the inventory range. Based on the demand response rules for interruptible loads, the interruption period range of the belt conveyor system is determined, and based on the interruption period range, the initial value of the interruption start time of the belt conveyor system is randomly determined. The interruptible load prediction model is specifically as follows: In the formula, This refers to the transfer volume data of the belt conveyor system at the j-th sampling time. This refers to the production and transportation volume data of the belt conveyor system at sampling time j. This represents the initial inventory level of the first silo. K represents the initial inventory of the second silo. th Nd represents the number of training samples for the model, and Nd represents the number of days. The width of the Gaussian kernel function. and It is a weight vector. and The bias coefficient, These are model learning samples for a single work scenario. T is a set of model learning samples for multiple single working scenarios. start This is the starting time for the daily interruptible load response of the transportation system.

5. The interruptible load control device based on a belt conveyor system according to claim 4, characterized in that, The day-ahead optimization scheduling model is specifically as follows: In the formula, and T represents respectively start The material levels in the first and second silos at any given time, wherein the first silo is located upstream of the second silo. and For balance coefficient, Let J be the real-time inventory of the first silo at sampling time j. This represents the real-time inventory of the second silo at sampling time j. This is the lower limit of the inventory in the first silo. This represents the maximum storage capacity of the first silo. This is the lower limit of the inventory in the second silo. This is the upper limit of the storage capacity of the second silo. Let j be the transport volume of the conveyor in the belt conveyor system at sampling time j. These represent the lower and upper limits of the conveyor's transport capacity in the belt conveyor system, respectively. Let $\frac{i}{j}$ be the belt speed of conveyor $i$ in the belt conveyor system at sampling time $j$. These are the lower and upper limits of the belt speed of conveyor i in the belt conveyor system, respectively. Let the mass of material per unit length of conveyor belt on conveyor i be the mass of material on conveyor i at sampling time j. This represents the upper limit of the material mass per unit length of the conveyor belt of conveyor i.

6. The interruptible load control device based on a belt conveyor system according to claim 5, characterized in that, The day-ahead optimization scheduling model solving unit is specifically used for: By solving the day-ahead optimization scheduling model, the silo inventory of the belt conveyor system when the day-ahead optimization scheduling model reaches the optimal solution is obtained; Based on the first silo inventory output by the optimized scheduling model, and combined with the relationship between the transfer volume of the first silo and the first silo inventory, the first maximum interruption end time is calculated. The critical point of the first maximum interruption end time is the longest time that the first silo can receive transfer volume after the belt conveyor system is interrupted. Based on the second silo inventory output by the previous day's optimized scheduling model, and combined with the relationship between the production and transportation volume of the second silo and the second silo inventory, the second maximum interruption end time is calculated. The second maximum interruption end time is the longest time that the inventory of the second silo can support the production process after the belt conveyor system is interrupted. The smaller of the first maximum interruption end time and the second maximum interruption end time is taken as the maximum interruption end time of the belt conveyor system in a single working scenario.

Images