Day-ahead rolling production scheduling method for central heating network and related equipment

Abstract

The present application belongs to the technical field of central heating network scheduling, and particularly relates to a day-ahead rolling production scheduling method for central heating network and related equipment, comprising: obtaining predicted meteorological data and inputting a building thermodynamics model to calculate the non-steady state on-demand load prediction sequence of each heat exchange station; calculating the dynamic transmission time lag according to the real-time flow rate integral; combining the time lag to generate an initial production scheduling sequence by reverse translation and superposition of the load prediction sequence; constructing a multi-objective optimization function containing a step attenuation penalty operator and a heat storage constraint condition containing a virtual heat storage state transition equation; inputting the initial production scheduling sequence into the solution to generate a smooth production scheduling control sequence, and issuing it to the source end executing mechanism. The present application can accurately align the space-time dislocation of supply and demand, and smooth the mutation load command at the source, effectively preventing water hammer effect and bottom equipment loss.

Description

Technical Field

[0001] This invention belongs to the field of centralized heating network scheduling technology, specifically involving a day-ahead rolling production scheduling method and related equipment for centralized heating network production scheduling. Background Technology

[0002] In the process of intelligent transformation of centralized heating networks, formulating day-ahead heating production plans based on outdoor weather changes and coordinating frequency control with main pump units is a conventional method to ensure heating demand and achieve energy conservation and carbon reduction in production. Currently, the industry's commonly used production plan formulation methods mainly rely on the comprehensive experience of dispatchers or issue macro-level dispatch instructions based on fixed static climate compensation curves. Meanwhile, when facing complex hydraulic conditions and heat transmission demands in the heating network, to prevent water hammer effects triggered by large-scale changes in the production plan at the heat source end, the mainstream control strategy is usually to enforce mechanical speed control on the downstream hardware side of the heating network through local edge gateways, or to perform secondary trimming of the issued control instructions, thereby achieving physical safety protection for the network fluid and underlying actuators.

[0003] However, the aforementioned existing technologies have significant limitations in actual operation. First, scheduling modes based on static curves or manual experience cannot quantify the unsteady heat dissipation mechanisms on the user side caused by drastic weather fluctuations, resulting in insufficient accuracy in end-point heat load prediction. Second, the flow of heat medium in long-distance transmission and distribution networks involves dynamic hourly transmission delays. Due to the lack of online and accurate time-delay inversion methods, the timing of the ignition and temperature rise at the heat source end often results in a severe spatiotemporal misalignment with the actual heat demand at the end, easily leading to periodic large-scale oversupply or undersupply in the heating area. Furthermore, the "downstream patching" control logic of forcibly implementing mechanical speed control or instruction pruning on the downstream hardware side not only significantly increases the communication and computing power consumption of the distributed gateway, but also accelerates the fatigue wear of physical assets such as electric valves and high-power pump sets when dealing with drastic changes in production scheduling instructions, severely restricting the long-term stable and efficient operation of the heating system at the global level. Summary of the Invention

[0004] The technical problem to be solved by this invention is to address the shortcomings of the prior art by providing a day-ahead rolling scheduling method and related equipment for centralized heating network production scheduling. This method is used to solve the technical problems of poor load forecasting accuracy caused by static experience-based scheduling in existing centralized heating systems, spatiotemporal misalignment of supply and demand caused by dynamic time delays in long-distance pipelines, and increased wear and tear on mechanical equipment such as valves and pumps caused by forced instruction trimming on the downstream hardware side to prevent water hammer.

[0005] The objective of this invention is achieved through the following technical solutions: In a first aspect, the present invention provides a day-ahead rolling production scheduling method for centralized heating network production scheduling, comprising: Acquire the predicted meteorological data for the target time period, input the predicted meteorological data into a preset building thermodynamics model, and calculate the unsteady on-demand load prediction sequence for each heat exchange station node in the target heating network; The real-time flow velocity of the main branch of the target heating network is obtained, and the characteristic line integral is performed based on the real-time flow velocity of the main branch to calculate the dynamic transmission delay of the heat medium to each heat exchange station node. Based on the dynamic transmission delay, the unsteady on-demand load prediction sequence of each heat exchange station node is reverse-time shifted and superimposed on the time axis to generate an initial heat source production sequence. A multi-objective optimization function and thermal storage constraints are constructed. The multi-objective optimization function includes a step decay penalty operator for limiting the difference in production scheduling between adjacent time steps. The thermal storage constraints include a virtual thermal storage state transition equation constructed based on the physical properties of pipeline fluid. The initial heat source production sequence is input into the multi-objective optimization function and the thermal storage constraint condition for closed-loop solution to generate a smooth heat source production control sequence. The smooth heat source scheduling control sequence is sent to the source-end actuator for heating operation.

[0006] As a further improvement of the present invention, the predicted meteorological data includes outdoor ambient temperature prediction time series. With solar radiation heat flux The building thermodynamic model is a first-order thermal resistance-heat capacity energy conservation model of indoor air and exterior walls; Based on the building thermodynamic model, the day-ahead number in the unsteady on-demand load forecast sequence is calculated. Hour, the Predicted heat load for each branch building The formula is:

[0007] In the formula, To set the indoor temperature constant; Let be the equivalent thermal resistance between the inner surface of the j-th building and the core layer of the wall. The preset thermal resistance of the exterior wall; The equivalent physical heat capacity of the porous medium in the exterior wall structure of building j; Calculate the step size for production scheduling; The coefficient of solar radiation transmission heat absorption rate of building exterior windows; Let be the equivalent effective heat-receiving area of ​​the glass windows on the sunny side of building j.

[0008] As a further improvement of the present invention, the dynamic transmission delay of the heat medium to each heat exchange station node is calculated by integrating the characteristic lines based on the real-time flow velocity of the main branch. Specifically, the dynamic transmission delay corresponding to the j-th heat exchange station is obtained by the following implicit integral equation. :

[0009] In the formula, Indicates at time Pipeline axial position The transient axial velocity of the fluid at the location is determined based on the real-time velocity of the main branch; This represents the axial physical distance from the first heat source station to the j-th heat exchange station.

[0010] As a further improvement of the present invention, the step of reverse time-shifting and superimposing the unsteady-state on-demand load prediction sequences of each heat exchange station node on the time axis according to the dynamic transmission delay to generate an initial heat source scheduling sequence includes: Combining fluid properties and pipe characteristics, the initial heat source is calculated. The heating power constantly injected into the pipeline network :

[0011] The heating power corresponding to each heat exchange station is time-aligned and superimposed using a variable time-delay transfer matrix to generate the initial heat source production sequence. Total power at time :

[0012] In the formula, For the day-ahead number in the unsteady-state on-demand load forecast sequence Hour, the The predicted heat load required for each branch building The actual working temperature and density of the heat transfer medium water. For the Kronecker-Dirac function operator, The comprehensive unsteady-state heat transfer coefficient of the insulation layer for long-distance pipelines. The specific heat capacity of water at constant pressure. The nominal inner diameter of the main power transmission and distribution line. For at any time Pipeline axial position The transient axial velocity of the fluid at that location.

[0013] As a further improvement of the present invention, the step decay penalty operator is a fourth-order higher-order step incremental decay operator, and the multi-objective optimization function... The specific formula is constructed as follows:

[0014] in The first in the production scheduling sequence The heating capacity output of the primary heat source station per hour; The initial heat source production sequence is in the 1st... Total power per hour; This is the nonlinear combustion energy consumption mapping function for a heat source boiler. This refers to the maximum permissible unsteady-state hourly ramp power amplitude threshold for heat source equipment. These are the preset weighted balance coefficients; This refers to the fourth-order higher-order step increment decay operator.

[0015] As a further improvement of the present invention, the thermal storage constraint condition is as follows:

[0016] in, The hard red line threshold represents the maximum permissible safe landslide rate for a single heat source unit; For the first The virtual thermal energy storage value of the fluid inside the hourly pipeline network; For the first The sum of the actual instantaneous heat absorption power conducted back to the source end from the end of the hourly pipeline network; This is the ceiling threshold for the maximum safe heat storage capacity of the pipeline network.

[0017] As a further improvement of the present invention, the step of inputting the initial heat source production sequence into the multi-objective optimization function and the thermal storage constraint conditions for closed-loop solution to generate a smooth heat source production control sequence includes: Using the Sequential Quadratic Programming (SQP) algorithm, the multi-objective optimization function is subjected to closed-loop iterative calculation in the target time domain within the solution space defined by constraints. The output power sequence that minimizes the objective function is used as the smooth heat source production scheduling control sequence.

[0018] Secondly, the present invention provides a day-ahead rolling production scheduling system for centralized heating network production scheduling, comprising: The load forecasting module is used to acquire forecast meteorological data within a target time period, input the forecast meteorological data into a preset building thermodynamics model, and calculate the unsteady on-demand load forecasting sequence for each heat exchange station node in the target heating network. The time-delay alignment module is used to receive the unsteady-state on-demand load prediction sequence; obtain the real-time flow velocity of the main branch of the target heating network; perform characteristic line integration based on the real-time flow velocity of the main branch to calculate the dynamic transmission time delay of the heat medium to each heat exchange station node; and, based on the dynamic transmission time delay, perform reverse time shifting and superposition of the unsteady-state on-demand load prediction sequence on the time axis to generate an initial heat source scheduling sequence. An optimization and smoothing module is used to receive the initial heat source production schedule sequence; construct a multi-objective optimization function and thermal storage constraints, wherein the multi-objective optimization function includes a step decay penalty operator for limiting the difference in production schedule between adjacent time steps, and the thermal storage constraints include a virtual thermal storage state transition equation constructed based on the physical properties of the pipeline fluid; input the initial heat source production schedule sequence into the multi-objective optimization function and the thermal storage constraints for closed-loop solution to generate a smoothed heat source production schedule control sequence; The execution control module is used to receive the smooth heat source scheduling control sequence and send the smooth heat source scheduling control sequence to the source-end actuator for heating operation.

[0019] Thirdly, the present invention provides a computer device, comprising: a processor and a computer-readable storage medium; A processor, adapted to execute computer programs; A computer-readable storage medium storing a computer program that, when executed by the processor, implements the day-ahead rolling production scheduling method for centralized heating network production scheduling as described above.

[0020] Fourthly, the present invention provides a computer-readable storage medium storing a computer program adapted to be loaded by a processor and executed as described above for the day-ahead rolling production scheduling method for centralized heating network production.

[0021] The beneficial effects of this invention are as follows: This invention provides a day-ahead rolling scheduling method for centralized heating network production scheduling. By inputting predicted meteorological data into a preset building thermodynamics model to calculate the unsteady-state on-demand load prediction sequence, it quantifies the unsteady heat dissipation mechanism on the load side under the influence of the outdoor environment, overcomes the blind adjustment limitation of static climate compensation curves, and clarifies precise heating demand. By collecting real-time flow velocity data from main branches and performing characteristic line integration to calculate dynamic transmission delay, and by generating an initial scheduling sequence through reverse time translation and superposition on the time axis, it accurately eliminates the delay effect caused by fluid transport in long-distance pipelines, enabling a high degree of time axis alignment between heat source supply nodes and terminal consumption nodes. By injecting a stepped attenuation penalty operator into the multi-objective optimization function, when the load difference between adjacent time steps is too large, it enforces energy amplitude limiting through a mathematical model, avoiding the generation of peak step commands. By using a virtual heat storage state transition equation constructed based on the physical properties of pipeline fluid as a constraint condition for closed-loop solution, it can actively convert the pipeline fluid volume into an elastic heat storage body and issue progressive heat power commands in advance to achieve pre-charging of the pipeline network. The complementary and synergistic nature of the various technical features elevates the smoothing process of instructions entirely to the computational source, generating a smooth and flexible production scheduling control sequence without spikes and jumps, which is then directly issued. This not only completely eliminates the secondary computational overhead and network latency caused by the mechanical speed control implemented by the downstream edge gateway, but also enables the underlying actuators to achieve uninterrupted pure mechanical execution within their own physical safety limits. This eliminates the potential disaster hazard of water hammer in the pipeline network caused by sudden flow changes, and at the same time greatly delays the fatigue wear cycle of electric valves and high-power pump sets in the pipeline network system. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention 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 some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a flowchart of the day-ahead rolling production scheduling method for centralized heating network production scheduling in an embodiment of the present invention; Figure 2 This is an internal structural diagram of a computer device in an embodiment of the present invention. Detailed Implementation

[0024] To make the objectives and technical solutions of this invention clearer and easier to understand, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. The specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention.

[0025] The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings and specific embodiments. The described embodiments are only some embodiments of the present invention, and not all embodiments.

[0026] Current centralized heating day-ahead scheduling largely relies on dispatchers' manual experience or static climate compensation curves, failing to quantify the unsteady heat dissipation caused by drastic weather fluctuations on the building side. Furthermore, the heat medium experiences hourly dynamic transmission delays in long-distance pipelines, leading to a time-space misalignment between heat source temperature increases and actual end-user heating demands, resulting in widespread oversupply and undersupply. Simultaneously, existing solutions employ downstream edge gateways for secondary command pruning and mechanical speed limiting to mitigate water hammer caused by large-scale scheduling command changes. This increases gateway communication and computing power consumption, and frequent valve and pump adjustments accelerate equipment fatigue and wear.

[0027] Example 1 This embodiment describes a day-ahead rolling production scheduling method for centralized heating network production scheduling, the process of which is as follows: Figure 1 As shown, step S1 is executed sequentially. S6: S1: Acquire forecast meteorological data for the target time period, input the forecast meteorological data into a preset building thermodynamics model, and calculate the unsteady-state on-demand load forecast sequence for each heat exchange station node in the target heating network. The execution entity is the central dispatch platform server of the heating network; the target time period is the day-ahead scheduling window for the next 24 hours; the calculation step size is set. h (discrete time label is) ).

[0028] The dispatch platform receives forecast meteorological data from weather forecasts, which includes hourly outdoor temperatures. hourly solar radiation heat flux The meteorological time series data is input into a preset building thermodynamics model to calculate the heat load for each branch building in the entire network. The data is then aggregated to obtain the unsteady-state on-demand load prediction sequence for each heat exchange station node. The complete 24-hour load time series for each heat exchange station node is output and passed to the next step.

[0029] Specifically, targeting the entire network's first For each independent zoned building, the unsteady baseline on-demand heat load forecast power for each time period is directly calculated using a first-order RC (thermal resistance-heat capacity) energy conservation model of indoor air and exterior walls. :

[0030] In the formula, To set the indoor temperature constant; Let be the equivalent thermal resistance between the inner surface of the j-th building and the core layer of the wall. The preset thermal resistance of the exterior wall; The equivalent physical heat capacity of the porous medium in the exterior wall structure of building j; Calculate the step size for production scheduling; The coefficient of solar radiation transmission heat absorption rate of building exterior windows; Let be the equivalent effective heat-receiving area of ​​the glass windows on the sunny side of building j.

[0031] S2: Obtain the real-time flow velocity of the main branch of the target heating network, perform characteristic line integration based on the real-time flow velocity of the main branch, and calculate the dynamic transmission delay of the heat medium to each heat exchange station node.

[0032] The scheduling platform reads real-time flow velocity and flow field data of the main branches collected by distributed IoT measuring points in the main network. Using the characteristic line integral method of the heat network fluid, for each heat exchange station, based on the distance from the heat source to the heat exchange station pipeline and the transient flow velocity at various positions along the pipeline axis, the dynamic transmission time delay of the heat medium from the heat source to the heat exchange station is implicitly integrated to solve. Output the hourly transmission delay of each heat exchange station as it dynamically changes with hydraulic conditions, and pass it on to the next step.

[0033] Specifically, the dynamic transmission delay of the heat medium reaching each heat exchange station node is calculated by integrating the characteristic lines based on the real-time flow velocities of the main branches. The dynamic transmission delay corresponding to the j-th heat exchange station is obtained through the following implicit integral equation. :

[0034] In the formula, Indicates at time Pipeline axial position The transient axial velocity of the fluid at the location is determined based on the real-time velocity of the main branch; This represents the axial physical distance from the first heat source station to the j-th heat exchange station. The flow velocity of the heat medium in the pipeline is not constant; the flow velocity varies at different locations and times. The characteristic line integral integrates the flow time along the actual flow path of the fluid particles, rather than simply dividing the distance by the average flow velocity, to achieve online calibration of the transmission time delay under variable flow conditions.

[0035] S3: Based on the dynamic transmission time delay, the unsteady on-demand load prediction sequence of each heat exchange station node is shifted backward on the time axis and superimposed to generate the initial heat source scheduling sequence.

[0036] Based on the dynamic transmission time delay corresponding to each heat exchange station As a shift, a reverse time-axis phase advance shift is performed on the unsteady-state on-demand load prediction sequence of the heat exchange station: that is, the heat load required at the end time t requires the heat source to... The corresponding energy is output at all times; at the same time, the heat loss along the long-distance pipeline is taken into account to correct the output power of the heat source; the power sequence of all heat exchange stations in the entire network after time delay translation and heat loss correction is superimposed and converged on the time axis of the heat source side to obtain the initial heat source production sequence without the application of ramp-up and heat storage constraints. The output is then passed to the next step.

[0037] Specifically, by combining fluid physical properties and pipeline attributes, the initial heat source is calculated at... The heating power constantly injected into the pipeline network :

[0038] The heating power corresponding to each heat exchange station is time-aligned and superimposed using a variable time-delay transfer matrix to generate the initial heat source production sequence. Total power at time :

[0039] In the formula, For the day-ahead number in the unsteady-state on-demand load forecast sequence Hour, the The predicted heat load required for each branch building The actual working temperature and density of the heat transfer medium water. For the Kronecker-Dirac function operator, The comprehensive unsteady-state heat transfer coefficient of the insulation layer for long-distance pipelines. The specific heat capacity of water at constant pressure. The nominal inner diameter of the main power transmission and distribution line. For at any time Pipeline axial position The transient axial velocity of the fluid at the location. The power of each heat exchange station after time delay offset is mapped back to the same time scale at the source end and summed to obtain the initial total power time series of the heat source without the application of ramp and heat storage constraints.

[0040] S4: Construct a multi-objective optimization function and thermal storage constraints. The multi-objective optimization function includes a step decay penalty operator for limiting the difference in production scheduling between adjacent time steps. The thermal storage constraints include a virtual thermal storage state transition equation constructed based on the physical properties of the pipeline fluid.

[0041] Using the hourly heating power of the heat source over the next 24 hours as the decision variable, a multi-objective optimization function is constructed. The optimization function has a built-in step decay penalty operator, which introduces a high penalty when the change in heat source power between adjacent hours is too large, suppressing the step jump of the command. At the same time, thermal storage constraints are constructed, including the virtual thermal storage state transition equation of the pipeline fluid, which is used to characterize the physical process of heat storage and heat release of large-volume heat medium in long-distance pipelines. The constraints also include the upper limit of the ramp-up and ramp-down rates of heat source equipment and the upper and lower limits of the thermal storage energy of the pipeline, thus fully defining the feasible solution space for optimization.

[0042] Specifically, the step decay penalty operator is a fourth-order higher-order step increment decay operator, and the multi-objective optimization function... The specific formula is constructed as follows:

[0043] in The first in the production scheduling sequence The heating capacity output of the primary heat source station per hour; The initial heat source production sequence is in the 1st... Total power per hour; This is the nonlinear combustion energy consumption mapping function for a heat source boiler. This refers to the maximum permissible unsteady-state hourly ramp power amplitude threshold for heat source equipment. These are the preset weighted balance coefficients; This refers to the fourth-order higher-order step increment decay operator.

[0044] The objective function minimizes the overall performance index while considering fuel cost, load tracking accuracy, and power fluctuation smoothness. When the power difference between adjacent hours approaches or even exceeds the equipment's allowable ramp-up threshold, the fourth-order operator outputs a penalty value that is rapidly amplified to suppress steep step jumps in the solution.

[0045] The thermal storage constraints are as follows:

[0046] in, The hard red line threshold represents the maximum permissible safe landslide rate for a single heat source unit; For the first The virtual thermal energy storage value of the fluid inside the hourly pipeline network; For the first The sum of the actual instantaneous heat absorption power conducted back to the source end from the end of the hourly pipeline network; This is the ceiling threshold for the maximum safe heat storage capacity of the pipeline network.

[0047] In this embodiment, the first row of constraints represents the upper and lower limits of the hourly power variation of the heat source equipment itself; the second row represents the virtual heat storage state transition equation of the pipeline network, where the current hourly heat storage = the heat storage at the previous moment + the energy injected by the heat source. Energy is extracted from the end of the entire network to depict the dynamic process of heat storage and release in the pipeline network; the third line shows that the heat storage in the pipeline network must not be less than zero or exceed the maximum safe heat storage limit to ensure hydraulic and thermal safety.

[0048] When a load spike is predicted in the future, the fourth-order operator does not allow the heat source to increase its power significantly in an instant. The optimization solver will allow the heat source to output more heat appropriately a few hours before the peak arrives, and the excess energy will be stored inside the pipeline fluid. At the peak moment, the stored heat in the pipeline will be released to buffer the sudden change in load and realize feedforward charging and heat storage. At the same time, all actions are bound within a safe range by the equipment ramp-up and the upper and lower limits of pipeline heat storage.

[0049] S5: Input the initial heat source production sequence into the multi-objective optimization function and the thermal storage constraint conditions for closed-loop solution to generate a smooth heat source production control sequence.

[0050] The initial heat source production sequence output by S3 is used as the reference target for optimized load tracking. The objective function and all constraints constructed by S4 are substituted into it, and a closed-loop iterative solution is performed in the 24-hour full time domain. After the solution is obtained, the optimal power time sequence that eliminates steep step peaks and presents a smooth parabolic shape is output, which is the smooth heat source production control sequence.

[0051] S6: Send the smooth heat source scheduling control sequence to the source-end actuator for heating operation.

[0052] The scheduling platform will smooth the heat source production control sequence and the corresponding pump group frequency conversion timing, and directly send the boiler controller, circulating pump group frequency converter and other source-end actuators at the first heat source station; the source-end actuators will directly complete the mechanical drive execution according to the sequence, without the need for the downstream heat exchange station edge gateway to perform secondary instruction trimming or forced mechanical speed limiting, thus completing the day-ahead rolling production closed-loop execution.

[0053] Substituting the initial heat source production schedule into the multi-objective optimization function, and applying all thermal storage constraints, the Sequential Quadratic Programming (SQP) algorithm is used to perform closed-loop iterative calculations in the time domain 24 hours in advance. The SQP algorithm iteratively transforms the constrained optimization problem into solving a quadratic programming subproblem, continuously updating the heat source production power decision variables until the objective function converges. The output is the 24-hour power time series that minimizes the objective function, which is the smoothed heat source production control sequence.

[0054] Sequential quadratic programming (SQP) is suitable for handling continuous optimization problems involving nonlinear objective functions and equality / inequality constraints. It can iteratively search for the optimal production power sequence within the feasible solution domain defined by equipment ramp-up and pipeline thermal storage constraints.

[0055] In summary, this embodiment utilizes building thermodynamics to transform future weather changes into the building's actual unsteady-state heat demand, obtaining the load of each heat exchange station. Then, by combining the real-time flow velocity of the pipeline network with the inverse calculation of the heat medium delivery lag time at each station, the demand at the end is pushed back to the heat source time axis, compensating for pipeline transport delays and pipe heat dissipation to obtain the unconstrained original production schedule demand. Furthermore, through multi-objective optimization with a high-order penalty operator, when a sudden load jump occurs, the penalty term suppresses sudden changes in heat source power. The optimizer automatically activates virtual heat storage constraints for the pipeline fluid, pre-heating and storing heat in the pipeline network, using the network volume to buffer load spikes, transforming the step demand into a smooth parabolic power change. Finally, the generated smooth sequence is directly given to the source-end equipment for execution, avoiding sudden change instructions from the algorithm source and avoiding the patching-style processing logic of downstream gateways.

[0056] Example 2 Based on Example 1, a specific implementation scheme for a day-ahead rolling production scheduling method for centralized heating network production scheduling is provided.

[0057] Step 1: Meteorological Feedforward and Load Baseline Derivation The central dispatch platform, through the industrial internet data platform of the backbone network, automatically feeds forward to the multi-source meteorological forecast environmental matrix (including hourly temperature and light intensity) of each functional zone for the next 24 hours at the daily day-ahead dispatch node (e.g., 14:00 the previous day). The system calls the RC physical mechanism model to keep the temperature control target constant at 20°C, and directly calculates the non-steady-state baseline on-demand load forecast time-series flow of each heat exchange station node in the entire network online, establishing the underlying rigid target of production supply.

[0058] Step 2: Variable Flow Velocity Time Delay Analysis and Time Axis Phase Lead Slip The central server of the twin scheduling brain reads the current actual hydraulic velocity and flow field of the main trunk line and initiates calculation formula 2. The system runs characteristic line feature integrals along the geometric topological path of the main trunk line, and calibrates online the time delay sequence of heat medium transmission caused by velocity slip (e.g., the time delay at the farthest station A is 5.2 hours, and the time delay at the nearest station B is 1.5 hours). The scheduling platform then performs reverse "phase advance slip" on the time axis of the twin space: the heat demand at a certain meteorological change point at station A within the next 24 hours is advanced by 5.2 hours on the time axis; the demand at station B is advanced by 1.5 hours. After the advance shift of each branch, they are superimposed and converged on the time axis at the source end, and the initial heat source scheduling plan total power time sequence flow is directly deduced by inverse solving.

[0059] Step 3: Injection of Stepped Decay Operator and Settlement of Optimal Smooth Production Scheduling Plan To prevent unrealistically steep jumps in the initial planned total power due to the superposition of loads from multiple stations (which, if directly issued, would trigger water hammer and damage boilers), the central dispatching system activates calculation formula 3. The algorithm incorporates an "economic ramp-up step decay operator" and a "large-volume virtual thermal storage constraint of the entire pipeline network" into the feedforward optimization process. When an extreme cold wave is predicted, causing a sharp drop in demand in the initial production schedule at midnight, the high-order step operator instantly triggers a large energy efficiency penalty. The algorithm then automatically reverse-engineers the in-transit thermal storage volume of the pipeline fluid, controlling the heat source to begin a gradual, smooth, and slow temperature increase several hours in advance, thus "feedforward charging" the heat within the long-distance pipeline network. The SQP optimization module ultimately calculates an optimal day-ahead rolling production power control sequence command flow that completely smooths out any sudden spikes and exhibits a flexible, smooth parabolic characteristic, without any numerical dead zones.

[0060] Step 4: Fully automated, fully mechanical, and disturbance-free production drive execution The central control system directly distributes the calculated optimal day-ahead smooth production scheduling control sequence and the corresponding frequency conversion control timing of the network pump group to the source-end thermal power plant's primary heat station controller, the main heating valve group of the industrial waste heat station, and the large-scale network circulating frequency conversion pump group via the network control bus. Since the entire large-amplitude production scheduling control sequence has already completed the absorption of large inertia technology ramp-up and virtual heat storage relocation in the pipeline network at the source-end algorithm level, the day-ahead actuators are 100% directly mechanically driven during transient switching, and the gradient of flow and pressure fluctuations in the field pipelines naturally operates within the vibration-resistant physical safety red line. The entire network requires no manual secondary trimming or speed reduction of instructions from the local edge gateway, enabling the entire system to smoothly and flexibly transition to a globally high-efficiency production track, achieving automated closed-loop large-scale production without disturbance.

[0061] Example 3 This embodiment provides a day-ahead rolling production scheduling system for centralized heating network production scheduling, including: a load forecasting module, a time-delay alignment module, an optimization and smoothing module, and an execution control module; the modules interact with each other through the internal bus of the scheduling platform to complete time-series data exchange.

[0062] Load Forecasting Module: Acquires forecast meteorological data for the target time period, inputs the forecast meteorological data into a preset building thermodynamics model, calculates the unsteady-state on-demand load forecast sequence for each heat exchange station node in the target heating network, and outputs the unsteady-state on-demand load forecast sequence to the time-delay alignment module. The module can internally execute R... C. Heat load calculation formula.

[0063] The time-delay alignment module receives the unsteady-state on-demand load forecast sequence output by the load forecast module; acquires the real-time flow velocity of the main branches of the target heating network; calculates the dynamic transmission time delay of the heat medium reaching each heat exchange station node based on the characteristic line integral of the real-time flow velocity of the main branches; performs reverse time shifting and superposition on the load sequence according to the dynamic transmission time delay to generate an initial heat source scheduling sequence; and outputs the initial heat source scheduling sequence to the optimization and smoothing module. Internally, the module performs time-delay integration, heat dissipation correction, and time-delay superposition calculations.

[0064] The optimization and smoothing module receives the initial heat source scheduling sequence from the time-delay alignment module; constructs a multi-objective optimization function and thermal storage constraints, including a stepped decay penalty operator and a virtual thermal storage state transition equation for the pipeline fluid; inputs the initial heat source scheduling sequence into the optimization function and thermal storage constraints for closed-loop solution, generating a smoothed heat source scheduling control sequence; and outputs the smoothed heat source scheduling control sequence to the execution control module. Internally, the module includes an SQP solver, which executes the objective function, constraints, and iterative solution logic.

[0065] Execution control module: Receives the smoothed heat source scheduling control sequence output by the optimization and smoothing module; sends it to the source-end actuator of the first heat source station through the industrial network bus to complete the heating operation.

[0066] Each module can be deployed on the same scheduling server or in a distributed manner. Modules exchange timing data through inter-process communication or industrial Ethernet. Modules can be implemented in software programs or partially use hardware acceleration units such as FPGAs to perform matrix and optimization calculations.

[0067] Specific limitations regarding the day-ahead rolling production scheduling system for centralized heating networks can be found in the above-mentioned limitations on the day-ahead rolling production scheduling method for centralized heating networks; the corresponding technical effects are equivalent and will not be repeated here. Each module in the aforementioned day-ahead rolling production scheduling system for centralized heating networks can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0068] Example 3 Figure 2 An internal structural diagram of a computer device is shown in one embodiment. This computer device may specifically be a terminal or a server. Figure 2As shown, the computer device includes a processor, memory, network interface, display, camera, and input device connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a day-ahead rolling scheduling method for centralized heating network production scheduling. The display can be a liquid crystal display (LCD) or an e-ink display. The input device can be a touch layer covering the display, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0069] As will be understood by those skilled in the art, computer equipment Figure 2 The structure shown is merely a block diagram of a portion of the structure related to the present invention and does not constitute a limitation on the computer device to which the present invention is applied. A specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0070] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method described above.

[0071] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0072] In summary, the embodiments of this application provide a day-ahead rolling production scheduling method, system, computer equipment, and storage medium for centralized heating network production scheduling, through building first-order R The C-type thermal resistance and thermal capacity model is used to calculate the unsteady-state on-demand load prediction sequence for each heat exchange station; real-time flow velocity of the main network is collected, and the dynamic transmission time delay of each heat exchange station is solved by characteristic line integration; based on the dynamic time delay, the load sequence is reverse-time shifted and superimposed to compensate for network heat dissipation, resulting in the initial heat source production sequence; a multi-objective optimization function containing a fourth-order stepped attenuation penalty operator is constructed, and constraints including the virtual thermal storage state transition equation of the network fluid are established; the SQP sequential quadratic programming algorithm is used to perform constrained closed-loop solution on the initial production sequence to obtain a smoothed heat source production control sequence; the smoothed sequence is directly sent to the source-end actuators to complete the heating. The first-order R-type model of the building is used. The C-mechanism model characterizes the wall's thermal capacity inertia and solar heat gain, obtaining unsteady-state on-demand heat load; the characteristic line integral is used to calibrate the hourly dynamic transmission time delay of the variable flow network online, and reverse phase lead slip is performed to realize the source The system achieves spatiotemporal alignment of loads while compensating for heat loss along the pipeline. A fourth-order high-order stepped attenuation penalty operator is intrinsically introduced into the source-end day-ahead multi-objective optimization function to mathematically penalize power step jumps. The large-volume heat medium inside the long-distance pipeline is abstracted as a virtual heat storage object, and a state transition equation is established as a hard constraint. The scheduling algorithm actively utilizes the pipeline for feedforward energy charging and heat storage to buffer load peaks brought by cold waves. The entire process is completed within the source-end production scheduling algorithm, outputting a smooth control sequence that has eliminated peak jumps. The underlying actuators directly execute the mechanical commands without relying on secondary intervention from the downstream hardware side.

[0073] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on its differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments. It should be noted that the technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification.

[0074] The above-described embodiments are merely preferred embodiments of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various improvements and substitutions without departing from the principles of the present invention, and these improvements and substitutions should also be considered within the scope of protection of the present invention. Therefore, the scope of protection of this invention should be determined by the scope of the claims.

Claims

1. A day-ahead rolling production scheduling method for centralized heating network production scheduling, characterized in that, include: Acquire the predicted meteorological data for the target time period, input the predicted meteorological data into a preset building thermodynamics model, and calculate the unsteady on-demand load prediction sequence for each heat exchange station node in the target heating network; The real-time flow velocity of the main branch of the target heating network is obtained, and the characteristic line integral is performed based on the real-time flow velocity of the main branch to calculate the dynamic transmission delay of the heat medium to each heat exchange station node. Based on the dynamic transmission delay, the unsteady on-demand load prediction sequence of each heat exchange station node is reverse-time shifted and superimposed on the time axis to generate an initial heat source production sequence. A multi-objective optimization function and thermal storage constraints are constructed. The multi-objective optimization function includes a step decay penalty operator for limiting the difference in production scheduling between adjacent time steps. The thermal storage constraints include a virtual thermal storage state transition equation constructed based on the physical properties of pipeline fluid. The initial heat source production sequence is input into the multi-objective optimization function and the thermal storage constraint condition for closed-loop solution to generate a smooth heat source production control sequence. The smooth heat source scheduling control sequence is sent to the source-end actuator for heating operation.

2. The day-ahead rolling production scheduling method for centralized heating network production scheduling according to claim 1, characterized in that, The predicted meteorological data includes the outdoor ambient temperature prediction time series. With solar radiation heat flux The building thermodynamic model is a first-order thermal resistance-heat capacity energy conservation model of indoor air and exterior walls; Based on the building thermodynamic model, the day-ahead number in the unsteady on-demand load forecast sequence is calculated. Hour, the Predicted heat load for each branch building The formula is: In the formula, To set the indoor temperature constant; Let be the equivalent thermal resistance between the inner surface of the j-th building and the core layer of the wall. The preset thermal resistance of the exterior wall; The equivalent physical heat capacity of the porous medium in the exterior wall structure of building j; Calculate the step size for production scheduling; The coefficient of solar radiation transmission heat absorption rate of building exterior windows; Let be the equivalent effective heat-receiving area of ​​the glass windows on the sunny side of building j.

3. The day-ahead rolling production scheduling method for centralized heating network production scheduling according to claim 1, characterized in that, The dynamic transport time delay of the heat medium to each heat exchange station node is calculated by integrating the characteristic lines based on the real-time flow velocity of the main branch. Specifically, the dynamic transport time delay corresponding to the j-th heat exchange station is obtained through the following implicit integral equation. : In the formula, Indicates at time Pipeline axial position The transient axial velocity of the fluid at the location is determined based on the real-time velocity of the main branch; This represents the axial physical distance from the first heat source station to the j-th heat exchange station.

4. The day-ahead rolling production scheduling method for centralized heating network production scheduling according to claim 3, characterized in that, The step of generating an initial heat source production sequence by reverse time shifting and superimposing the unsteady-state on-demand load prediction sequences of each heat exchange station node on the time axis based on the dynamic transmission delay includes: Combining fluid properties and pipe characteristics, the initial heat source is calculated. The heating power constantly injected into the pipeline network : The heating power corresponding to each heat exchange station is time-aligned and superimposed using a variable time-delay transfer matrix to generate the initial heat source production sequence. Total power at time : In the formula, For the day-ahead number in the unsteady-state on-demand load forecast sequence Hour, the The predicted heat load required for each branch building The actual working temperature and density of the heat transfer medium water. For the Kronecker-Dirac function operator, The comprehensive unsteady-state heat transfer coefficient of the insulation layer for long-distance pipelines. The specific heat capacity of water at constant pressure. The nominal inner diameter of the main power transmission and distribution line. For at any time Pipeline axial position The transient axial velocity of the fluid at that location.

5. The day-ahead rolling production scheduling method for centralized heating network production scheduling according to claim 1, characterized in that, The stepped decay penalty operator is a fourth-order higher-order stepped incremental decay operator, and the multi-objective optimization function... The specific formula is constructed as follows: in The first in the production scheduling sequence The heating capacity output of the primary heat source station per hour; The initial heat source production sequence is in the 1st... Total power per hour; This is the nonlinear combustion energy consumption mapping function for a heat source boiler. This refers to the maximum permissible unsteady-state hourly ramp power amplitude threshold for heat source equipment. These are the preset weighted balance coefficients; This refers to the fourth-order higher-order step increment decay operator.

6. The day-ahead rolling production scheduling method for centralized heating network production scheduling according to claim 5, characterized in that, The thermal storage constraint condition is as follows: in, The hard red line threshold represents the maximum permissible safe landslide rate for a single heat source unit; For the first The virtual thermal energy storage value of the fluid inside the hourly pipeline network; For the first The sum of the actual instantaneous heat absorption power conducted back to the source end from the end of the hourly pipeline network; This is the ceiling threshold for the maximum safe heat storage capacity of the pipeline network.

7. The day-ahead rolling production scheduling method for centralized heating network production scheduling according to claim 1, characterized in that, The step of inputting the initial heat source production schedule sequence into the multi-objective optimization function and the thermal storage constraints for closed-loop solution to generate a smooth heat source production schedule control sequence includes: Using the Sequential Quadratic Programming (SQP) algorithm, the multi-objective optimization function is subjected to closed-loop iterative calculation in the target time domain within the solution space defined by constraints. The output power sequence that minimizes the objective function is used as the smooth heat source production scheduling control sequence.

8. A day-ahead rolling production scheduling system for centralized heating network production, characterized in that, include: The load forecasting module is used to acquire forecast meteorological data within a target time period, input the forecast meteorological data into a preset building thermodynamics model, and calculate the unsteady on-demand load forecasting sequence for each heat exchange station node in the target heating network. The time-delay alignment module is used to receive the unsteady-state on-demand load prediction sequence; obtain the real-time flow velocity of the main branch of the target heating network; perform characteristic line integration based on the real-time flow velocity of the main branch to calculate the dynamic transmission time delay of the heat medium to each heat exchange station node; and, based on the dynamic transmission time delay, perform reverse time shifting and superposition of the unsteady-state on-demand load prediction sequence on the time axis to generate an initial heat source scheduling sequence. An optimization and smoothing module is used to receive the initial heat source production schedule sequence; construct a multi-objective optimization function and thermal storage constraints, wherein the multi-objective optimization function includes a step decay penalty operator for limiting the difference in production schedule between adjacent time steps, and the thermal storage constraints include a virtual thermal storage state transition equation constructed based on the physical properties of the pipeline fluid; input the initial heat source production schedule sequence into the multi-objective optimization function and the thermal storage constraints for closed-loop solution to generate a smoothed heat source production schedule control sequence; The execution control module is used to receive the smooth heat source scheduling control sequence and send the smooth heat source scheduling control sequence to the source-end actuator for heating operation.

9. A computer device, characterized in that, include: Processor and computer-readable storage media; A processor, adapted to execute computer programs; A computer-readable storage medium storing a computer program that, when executed by the processor, implements the day-ahead rolling production scheduling method for centralized heating network production scheduling as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program adapted to be loaded by a processor and executed by the day-ahead rolling production scheduling method for centralized heating network production scheduling as described in any one of claims 1 to 7.

Images