Control method and device of heating system, electronic equipment and storage medium

Abstract

The application provides a control method and device of a heating system, an electronic device and a storage medium; the method comprises: acquiring a first working parameter of each first device in the heating system and a historical heat load; predicting a predicted heat load of the heating system based on the historical heat load, wherein the predicted heat load is a heat load in a preset time period in the future of the heating system; performing feature extraction on the first working parameter of each first device to obtain a first feature vector of each first device; performing mapping processing on the first feature vector of each first device respectively to obtain a health index of each first device; generating a constraint condition of a control amount of the heating system based on the health index; predicting a predicted control amount meeting the constraint condition based on the health index and the predicted heat load; generating a control instruction based on the predicted control amount, and controlling the heating system based on the control instruction. Through the application, the safety of the heating system control can be improved and energy consumption can be saved.

Description

Technical Field

[0001] This application relates to artificial intelligence and heating system technology, and in particular to a control method, device, electronic equipment and storage medium for a heating system. Background Technology

[0002] Heating systems deliver heat to users through heat sources, pipe networks, and circulating pumps. Their control objectives aim to balance user heating needs (such as stable room temperature), reduced operating energy consumption, and equipment safety. However, in actual operation, related technologies often fail to integrate control strategies with equipment health management (PHM). Circulating pumps, boilers, and other equipment experience performance degradation due to wear, aging, or scaling over long-term operation. Existing control algorithms typically assume ideal equipment conditions, focusing only on load matching and energy saving, neglecting equipment health. This can lead to heating systems issuing high-load commands to "sub-healthy" equipment, accelerating damage and even causing unplanned shutdowns. Secondly, there is a lack of physical safety boundary constraints. In pursuit of rapid response or extreme energy savings, optimization algorithms may output aggressive control commands (such as drastic temperature changes or excessively low pressure), potentially causing safety hazards such as pump cavitation, boiler thermal shock, or pipe network overpressure. Currently, there are no effective methods to improve the safety and energy efficiency of heating systems. Summary of the Invention

[0003] This application provides a control method, device, electronic equipment, and storage medium for a heating system, which can improve the safety of heating system control and save energy.

[0004] The technical solution of this application embodiment is implemented as follows: This application provides a control method for a heating system, the method comprising: Obtain the first operating parameters and historical heat load of each primary device in the heating system; Based on the historical heat load, the predicted heat load of the heating system is predicted, wherein the predicted heat load is the heat load of the heating system within a preset time period in the future; Feature extraction is performed on the first operating parameters of each first device to obtain a first feature vector for each first device; Based on the first feature vector of each of the first devices, a mapping process is performed to obtain the health index of each of the first devices, wherein the health index is a parameter used to measure the performance status of the first devices; Constraints for generating control quantities of the heating system based on the health index; Based on the health index and the predicted heat load, predictive control quantities that satisfy the constraints are predicted. Control commands are generated based on the predicted control values, and the heating system is controlled based on the control commands.

[0005] This application provides a control device for a heating system, including: The data acquisition module is used to acquire the first operating parameters of each primary device in the heating system, as well as the historical heat load; The load prediction module is used to predict the predicted heat load of the heating system based on the historical heat load, wherein the predicted heat load is the heat load of the heating system within a preset time period in the future. An index prediction module is used to extract features from the first operating parameters of each of the first devices to obtain a first feature vector for each of the first devices; perform mapping processing on the first feature vector of each of the first devices to obtain a health index for each of the first devices, wherein the health index is a parameter used to measure the performance status of the first devices; and generate control constraints for the heating system based on the health index. The control module is used to predict the predictive control quantity that satisfies the constraints based on the health index and the predicted heat load; generate control commands based on the predictive control quantity; and control the heating system based on the control commands.

[0006] This application provides an electronic device, the electronic device comprising: Memory is used to store executable instructions or computer programs. The processor, when executing computer-executable instructions or computer programs stored in the memory, implements the control method of the heating system provided in the embodiments of this application.

[0007] This application provides a computer-readable storage medium storing computer-executable instructions or computer programs, which, when executed by a processor, implement the control method of the heating system provided in this application.

[0008] This application provides a computer program product, including a computer program or computer-executable instructions. When the computer program or computer-executable instructions are executed by a processor, they implement the control method of the heating system provided in this application.

[0009] The embodiments of this application have the following beneficial effects: By acquiring the first operating parameters and historical heat load of the first device, and predicting the heat load and calculating the device's health index accordingly, dynamic constraints are generated based on the health index. Under these constraints, control quantities that meet the heat load requirements are predicted, and finally, control commands are generated to control the system. This achieves deep coupling between the operation control of the heating system and the health status of the equipment. By extracting features from the first operating parameters and mapping them to obtain the health index, the performance status of the equipment in the heating system can be quantified and perceived in real time. Constraints for the control quantities are then dynamically generated based on this health index, ensuring that the generated predicted control quantities always remain within the current allowable operating range of the equipment. This mechanism ensures that the control commands can adaptively match the actual health status of the equipment, thereby effectively delaying the wear and tear process of equipment in a performance degradation state and extending the service life of the first device while ensuring heating demand. It also enhances the global optimization capability of the heating system under complex operating conditions. By introducing the predicted heat load within a preset future time frame based on historical load prediction, the predicted heat load represents the heating demand that the heating system needs to meet. Using the predicted heat load and the equipment's health index together as the basis for calculating the predicted control quantities, the optimal control quantities can be solved under the dual premise of meeting future heat demand and equipment health constraints. This combination of feedforward prediction and health constraints enables the control system to adjust in advance before load fluctuations occur and ensures the health of the equipment, thereby achieving a synergistic improvement in heating quality assurance and overall operating efficiency of the heating system. Attached Figure Description

[0010] Figure 1 This is a schematic diagram illustrating the application mode of the control method for the heating system provided in the embodiments of this application; Figure 2 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application; Figure 3A This is a first flowchart illustrating the control method of the heating system provided in the embodiments of this application; Figure 3B This is a second flowchart illustrating the control method of the heating system provided in the embodiments of this application; Figure 4 This is a third flowchart illustrating the control method for a heating system provided in an embodiment of this application; Figure 5 This is a structural diagram of the control device for the heating system provided in the embodiments of this application.

[0011] It should be noted that the terms "first" and "second" mentioned above are only used to distinguish between different options and do not represent the degree of superiority or inferiority of the options or their priority in the implementation process. Detailed Implementation

[0012] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0013] It should be noted that the data collection and processing in this application should strictly comply with the requirements of relevant laws and regulations, obtain the informed consent or separate consent of the personal information subject, and carry out subsequent data use and processing within the scope of laws and regulations and the authorization of the personal information subject.

[0014] In the embodiments of this application, the terms "module" or "unit" refer to a computer program or part of a computer program that has a predetermined function and works with other related parts to achieve a predetermined goal, and can be implemented wholly or partially using software, hardware (such as processing circuitry or memory), or a combination thereof. Similarly, a processor (or multiple processors or memory) can be used to implement one or more modules or units. Furthermore, each module or unit can be part of an overall module or unit that includes the functionality of that module or unit.

[0015] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0016] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0017] In the following description, the terms "first, second, third" are used merely to distinguish similar objects and do not represent a specific ordering of objects. It is understood that "first, second, third" may be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0018] Before providing a further detailed description of the embodiments of this application, the nouns and terms involved in the embodiments of this application will be explained, and the nouns and terms involved in the embodiments of this application shall be interpreted as follows.

[0019] 1) Heating System: This is a comprehensive technical system mainly composed of boiler room units, outdoor heating pipe networks, and radiators. It achieves heat energy transmission and distribution through the circulation of a heat medium. Heat source equipment includes coal-fired boilers, gas-fired boilers, and clean energy heat pumps, etc. Heat transport equipment relies on pipelines, water pumps, etc., to form a heat network. Terminal heat dissipation equipment adopts radiators or underfloor heating. Central heating systems typically use hot water or steam as the medium.

[0020] 2) Heat Load: The amount of heat supplied to a building by the heating system. The design heat load of a heating system refers to the amount of heat supplied to a building per unit time to achieve the required indoor temperature under the outdoor design temperature. Heat load is the most fundamental data for designing a heating system. According to the basic theory of heat transfer, to maintain a certain indoor temperature, the heat gain and heat loss in the room must be balanced. The calculation method for the design heat load of a heating system is determined based on the principle of heat balance.

[0021] 3) Health Index (HI): This is a quantitative indicator used to assess the current health status of equipment or systems. It is usually expressed as a continuous value, such as from 0 to 1 or a percentage range, where 1 represents good health. A health index below a threshold reflects equipment performance degradation or potential failure risk.

[0022] In the actual operation of heating systems, there is a significant contradiction between achieving control objectives and the actual operating conditions of equipment. Related technologies face the following major challenges: The contradiction between load fluctuations and large thermal inertia makes precise control difficult. Heating loads exhibit dynamic fluctuations due to outdoor meteorological conditions (temperature, wind speed, sunlight) and user-side heating behavior, while the heating network exhibits significant thermal inertia and transmission lag. Traditional feedback control (such as PID) often has a delayed response, leading to decreased heating quality or energy waste. Although existing model predictive control (MPC) technology can incorporate load forecasting for feedforward regulation, it is usually based on idealized equipment models, ignoring the heterogeneity and time-varying nature of the physical system. Control strategies are severely disconnected from equipment health status (PHM). During long-term operation, key equipment in the system (such as circulating pumps, boilers, heat exchangers, and regulating valves) inevitably experience varying degrees of performance degradation, such as bearing wear, impeller cavitation, flow channel scaling, or mechanical aging. Existing control systems typically assume that actuators have rated performance, calculating control commands only with load matching and optimal energy efficiency as objectives. Existing Predictive Fault and Health Management (PHM) systems typically operate as independent monitoring units, possessing only alarm functions and unable to directly participate in closed-loop control. This disconnect means that when equipment is in a "sub-healthy" state, the control system may still issue high-load or highly dynamic commands, accelerating equipment wear and tear, and even triggering unplanned shutdowns. There is also a conflict between operational safety and the goal of extreme optimization. With the application of intelligent optimization algorithms, control systems tend to pursue extreme energy consumption indicators or response speeds, potentially outputting overly aggressive control quantities (such as extremely rapid water temperature rise and fall rates or extremely low pump inlet pressure). Without real-time safety boundary constraints based on physical mechanisms, such commands may cause system parameters to exceed limits, leading to serious safety accidents such as pipeline overpressure and pipe rupture, pump cavitation, or boiler thermal shock.

[0023] This application provides a control method, control device, electronic device, computer-readable storage medium, and computer program product for a heating system, which can improve the safety of heating system control and save energy.

[0024] The following describes exemplary applications of the electronic devices provided in the embodiments of this application. These electronic devices can be implemented as terminal devices, such as laptops, tablets, desktop computers, set-top boxes, smart TVs, in-vehicle terminals, virtual reality (VR) devices, augmented reality (AR) devices, and other various types of terminals. They can also be implemented as servers. The following will describe exemplary applications when the electronic device is implemented as a terminal device or a server.

[0025] refer to Figure 1 , Figure 1This is a schematic diagram illustrating the application mode of the control method for the heating system provided in the embodiments of this application; for example, Figure 1 The system involves server 200, network 300, and terminal devices (e.g., terminal devices 400-1 to 400-N). Terminal device 400 is connected to voice recognition server 200 through network 300. Network 300 can be a wide area network, a local area network, or a combination of both.

[0026] Terminal devices 400-1 to 400-N are control devices respectively installed on each device in the heating system (e.g., circulating pump / pipeline pump / primary network pump / secondary network pump / critical regulating valve / zoning valve / main valve). They are used to acquire the operating status data of each device and send the operating status data to server 200 via network 300. Server 200 is the control device of the heating system. Server 200 is used to invoke the heating system control method provided in the embodiments of this application, determine the health index of each device based on the operating status data, predict the control quantity of each device in the heating system based on the health index, generate control instructions carrying the control quantities, and send the control instructions to each terminal device to control each device in the heating system.

[0027] In some embodiments, the server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. Electronic devices can be smartphones, tablets, laptops, desktop computers, smart speakers, smartwatches, etc., but are not limited to these. Terminal devices and servers can be directly or indirectly connected via wired or wireless communication, which is not limited in this embodiment.

[0028] See Figure 2 , Figure 2 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device may be... Figure 1 Server 200, Figure 2 The server 200 shown includes at least one processor 410, memory 450, and at least one network interface 420. Each component in server 200 is coupled together via a bus system 440. It is understood that the bus system 440 is used to implement communication between these components. In addition to a data bus, the bus system 440 also includes a power bus, a control bus, and a status signal bus. However, for clarity, ... Figure 2 The general labeled all buses as Bus System 440.

[0029] Processor 410 can be an integrated circuit chip with signal processing capabilities, such as a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor can be a microprocessor or any conventional processor, etc.

[0030] The memory 450 may be removable, non-removable, or a combination thereof. Exemplary hardware devices include solid-state storage, hard disk drives, optical disk drives, etc. The memory 450 may optionally include one or more storage devices physically located away from the processor 410.

[0031] The memory 450 may include volatile memory or non-volatile memory, or both. The non-volatile memory may be read-only memory (ROM), and the volatile memory may be random access memory (RAM). The memory 450 described in this application embodiment is intended to include any suitable type of memory.

[0032] In some embodiments, memory 450 is capable of storing data to support various operations, examples of which include programs, modules, and data structures or subsets or supersets thereof, as illustrated below.

[0033] Operating system 451 includes system programs for handling various basic system services and performing hardware-related tasks, such as the framework layer, core library layer, driver layer, etc., for implementing various basic business functions and handling hardware-based tasks; The network communication module 452 is used to reach other electronic devices via one or more (wired or wireless) network interfaces 420, exemplary network interfaces 420 including: Bluetooth, WiFi, and Universal Serial Bus (USB), etc. In some embodiments, the apparatus provided in this application can be implemented in software. Figure 2 A control device 455 for a heating system, stored in memory 450, is shown. This device can be software in the form of programs and plug-ins, including the following software modules: a data acquisition module 4551, a load prediction module 4552, an index prediction module 4553, and a control module 4554. These modules are logically linked and can therefore be arbitrarily combined or further separated according to the functions they implement. Figure 2 For ease of explanation, all the modules mentioned above are shown at once, and the function of each module will be explained below.

[0034] In other embodiments, the control device for the heating system provided in this application can be implemented in hardware. As an example, the control device for the heating system provided in this application can be a processor in the form of a hardware decoding processor, which is programmed to execute the control method of the heating system provided in this application. For example, the processor in the form of a hardware decoding processor can be one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), or other electronic components.

[0035] The control method of the heating system provided in the embodiments of this application will be described in conjunction with the exemplary application and implementation of the server provided in the embodiments of this application.

[0036] The following describes the control method for the heating system provided in the embodiments of this application. As mentioned above, the electronic device implementing the control method for the heating system in the embodiments of this application can be a terminal device or a server, or a combination of both. Therefore, the executing entity of each step will not be described again below.

[0037] See Figure 3A , Figure 3A This is a flowchart illustrating the control method of the heating system provided in the embodiments of this application, which will be combined with... Figure 3A The steps shown are explained. Figure 3A The entity responsible for executing the steps is Figure 1 Server 200.

[0038] In step 301, the first operating parameters of each first device in the heating system and the historical heat load are obtained.

[0039] For example, the first type of equipment is equipment whose failure would directly lead to heating interruption, significant safety risks, or widespread failure to meet room temperature standards. Types of first-type equipment include pumps, heat source equipment, heat exchangers, and valves. Equipment other than the first type of equipment refers to devices in the heating system that do not affect system energy consumption or operating boundaries, such as auxiliary lighting and valves on terminal branches.

[0040] Pumps include: circulating pumps, pipeline pumps, primary network pumps, and secondary network pumps (including motors and bearings). Pumps directly determine flow rate and pressure difference, and are subject to risks such as cavitation, overload, and vibration degradation.

[0041] Heat source equipment includes: boilers, heat source units, and heat pump units. The heat source equipment determines the heating output and water supply temperature, as well as the risks of thermal shock, overheating, and overpressure.

[0042] Heat exchangers include plate heat exchangers (in-station heat exchange units), which affect heat exchange capacity and secondary network temperature supply.

[0043] Valves include: critical control valves, zone valves, and main valves. Valves affect the main flow distribution and pressure differential.

[0044] For example, the first operating parameters include: supply and return water temperatures, network pressure, pump frequency / current, valve opening, outdoor temperature, and zone heat load. Pump frequency typically refers to the power supply frequency of the motor driving the pump. Heat load is the amount of heat supplied to a building by the heating system. Historical heat load is the heat load of the heating system periodically collected over a preset period of time. For example, if data is collected every hour over the past six hours, six heat load values ​​are obtained as historical heat load, which can be represented as a sequence.

[0045] In step 302, the predicted heat load of the heating system is predicted based on historical heat load.

[0046] Here, the predicted heat load is the heat load of the heating system within a preset time period in the future.

[0047] For example, the forecasting process is implemented through a forecasting model, which can be a Long Short-Term Memory (LSTM) network combined with a Gradient Boosting Decision Tree (GBDT) model. To improve forecast accuracy, historical heat load and future predicted outdoor temperature can be used as inputs; the future predicted outdoor temperature can be obtained from the meteorological application platform.

[0048] In some embodiments, step 302 can be implemented as follows: obtaining the predicted outdoor temperature within a preset time period; performing feature extraction processing on a first prediction model based on historical heat load to obtain a first feature sequence, wherein the first feature sequence includes a second feature vector corresponding to the load at different times; performing feature extraction processing on the first prediction model based on the predicted outdoor temperature within a preset time period to obtain a second feature sequence, wherein the second feature sequence includes a temperature feature vector corresponding to the temperature at different times; performing iterative prediction processing on the first prediction model based on the first feature sequence and the second feature sequence to obtain multiple sub-predicted heat loads at different times, wherein the output of each iteration of the iterative prediction processing is the sub-predicted heat load at different times, and the input of each iteration of the iterative prediction processing is the already predicted sub-predicted heat load, the first feature sequence, and the second feature sequence; and combining each sub-predicted heat load into a predicted heat load in sequence form.

[0049] For example, the second characteristic sequence over the next N hours (N is a positive integer) is, for example, [T1, ..., TN], where T represents temperature. This is a predicted heat load in sequence form. It can be represented as ,in, This indicates the current time (either discrete sampling time or continuous time). This refers to the number of prediction steps or the length of the prediction time domain, such as predicting the next 1 to 6 hours. Each step length. () represents the range of a time interval. This is the actual heat load. Assuming the heat load is predicted for the next 1 to 6 hours, with a period of one hour, the input for the prediction processing corresponding to the heat load of the 5th hour includes: the sub-predicted heat loads corresponding to the 1st, 2nd, 3rd, and 4th hours respectively, the first feature sequence, and the second feature sequence.

[0050] In step 303, feature extraction is performed on the first operating parameters of each first device to obtain the first feature vector of each first device.

[0051] For example, feature extraction is implemented through a neural network model. The first working parameters include multiple types, each type is extracted separately to obtain different types of features, and the different types of features are combined into a first feature vector.

[0052] In some embodiments, step 303 can be implemented by performing the following processing for each first device: extracting features for each first operating parameter of the first device to obtain multiple parameter features; and concatenating each parameter feature to obtain a first feature vector.

[0053] For example, based on the vibration, current, and temperature rise values ​​of the pump or motor, the neural network model is called sequentially to perform feature extraction processing to obtain the thermodynamic features of the temperature rise value, the electrical features of the current, the time-domain features and frequency-domain features of the motor vibration, and the thermodynamic features, electrical features, time-domain features and frequency-domain features are concatenated into the first feature vector.

[0054] In step 304, a mapping process is performed based on the first feature vector of each first device to obtain the health index of each first device.

[0055] Here, the health index is a parameter used to measure the performance status of the first device.

[0056] For example, the mapping process includes normalization and prediction. The results of the normalization and prediction processes are weighted and summed to obtain the health index.

[0057] In some embodiments, step 304 can be implemented by performing the following processing for each first device: calling a second prediction model based on a first feature vector to predict the probability that the first device is in a healthy state; normalizing the first feature vector to obtain a normalized value; and weighted summing the predicted probability and the normalized value to obtain a health index.

[0058] For example, the second prediction model can be a lightweight machine learning model. Based on multi-dimensional feature vectors, the lightweight machine learning model is called for prediction processing to obtain the predicted probability that the heating system is in a healthy state. Based on the vibration, current, and temperature rise values ​​of the pump or motor, the normalization function of the normalization layer is called. Preset physical limit standards (such as ISO 10816 vibration standard, motor insulation class temperature limit) are compared with the collected operating parameters. If the operating parameters are within the standard range, they are mapped to normalized values ​​according to the normalization formula. If the operating parameters are not within the standard range, they are mapped to 0. The predicted probability and the normalized value are weighted and summed to obtain the health index of the first device.

[0059] In step 305, constraints on the control quantities of the heating system are generated based on the health index.

[0060] For example, constraints are generated for each first device. The constraint for maximum pump frequency or maximum flow rate is expressed as follows (1.1): (1.1) The constraint condition for the maximum rate of ascent is expressed by the following formula (1.2): (1.2) Among them, the lower the health index, the smaller the rate of increase. The rate of increase (Maximum Ramp Rate), also known as the rate of change or oscillation rate, refers to the magnitude of change that a control system allows equipment parameters to undergo per unit time. When... When the value falls below a threshold, a candidate set for standby machine switching is generated (e.g., enabling a standby pump). It is the equipment number (number) (e.g., pumps / boilers / valves) It is a moment equipment The maximum permissible control quantity (such as maximum pump frequency / maximum opening / maximum output). It is the equipment's rated allowable control quantity (rated pump frequency / rated output, etc.); It is a mapping function from health index to traffic (a common example of monotonically non-increasing health index is that the lower the allowable upper limit). It is the change (increase or decrease) of the control quantity in adjacent control cycles. It is a mapping function from the health index to the upper limit of the rate of increase (the lower the health index, the slower the change).

[0061] In step 306, based on the health index and predicted heat load, a predictive control quantity that satisfies the constraints is predicted.

[0062] For example, an objective function is constructed based on the health index and predicted heat load. The objective function is used to simultaneously minimize energy consumption and deviation, and to penalize equipment wear when the health index decreases.

[0063] In some embodiments, reference Figure 3B , Figure 3B This is a second flowchart illustrating the control method of the heating system provided in this application embodiment; step 306 can be achieved through... Figure 3B Steps 3061 to 3063 are implemented, and the details are explained below.

[0064] In step 3061, the actual energy consumption parameters are calculated based on the predicted heat load.

[0065] For example, actual energy consumption parameters include total fixed energy consumption and actual water supply temperature, which can be calculated using a physical model of the heating system.

[0066] In step 3062, an objective function is constructed based on actual energy consumption parameters, health index, and control variables.

[0067] In some embodiments, the control variables include: a set water supply temperature value and equipment control parameters; in some embodiments, step 3062 can be implemented by: calculating the square norm of the difference between the actual water supply temperature value and the set water supply temperature value; calculating a first ratio between the sum of the equipment energy consumption of each first device and the total rated energy consumption; calculating a second ratio between the equipment control parameters of each first device and the health index, and performing a weighted summation on each second ratio to obtain a first weighted result; using the first weighted result, the first ratio, and the weighted summation of the square norm as the objective function.

[0068] For example, the objective function is expressed as formula (2) below, and the control quantity obtained based on the objective function satisfies the dynamic constraint conditions mentioned above: (2) Where J is the objective function, and the optimization objective is to minimize the value of the objective function; These are weighting coefficients (used to balance energy consumption, comfort / tracking error, and wear / health constraints); This refers to energy consumption, which can be specifically obtained for each piece of equipment through its respective decision variables. For example, boiler fuel consumption can be obtained based on boiler output, pump power consumption can be obtained based on pump frequency, and valve energy consumption can be obtained based on valve opening degree. It is the total fixed energy consumption, calculated based on the predicted heat load; It is the actual water supply temperature value, which can be the primary network or the secondary network water supply temperature, and is determined by predicting the heat load. It sets the water supply temperature value or target temperature trajectory; It is the square norm; , It is the first weighted result, which increases with decreasing health index (lower HI) and increasing load / control quantity ( (Increases) and increases. It is a control variable (which can be a vector such as pump frequency, valve opening, boiler output, etc.). It is the weight value; The value of the function increases as the health index decreases and the load increases.

[0069] In step 3063, the predictive control model is invoked to calculate the target value of the control variable that minimizes the value of the objective function and satisfies the constraints, and the target value is used as the predictive control quantity.

[0070] The objective function is used to calculate the predictive control input that minimizes the objective function for each device. The objective function can be solved using Model Predictive Control (MPC) to obtain the predictive control input for each key device.

[0071] Continue to refer to Figure 3A In step 307, control commands are generated based on the predicted control variables, and the heating system is controlled based on the control commands.

[0072] For example, control commands are sent to the programmable logic controller (PLC) of the heating system, so that the PLC writes the control commands into the frequency converter and valves to control the heating system.

[0073] In some embodiments, the "generating control instructions based on predictive control quantities" in step 307 can be implemented in the following way: performing calculations based on predictive control quantities and the physical model of the heating system to obtain the second operating parameters of the heating system; performing the following processing for each second operating parameter: detecting whether the second operating parameter is within a safe range; in response to the second operating parameter being within a safe range, determining the predictive control quantity associated with the second operating parameter; and generating the control instructions corresponding to the associated predictive control quantity.

[0074] For example, a physical model is a simplified expression that uses mathematical formulas and logical rules to describe a real-world physical object or process. Setting up a physical model involves safety checks on control quantities predicted by artificial intelligence to prevent these predicted control quantities from exceeding corresponding safety boundaries. After receiving a control command, the heating system executes the corresponding control command and generates corresponding operating parameters. For example, the control quantities carried by the control command include: pump frequency, valve opening, boiler output, and water supply temperature. The generated operating parameters include: pressure, temperature, flow rate, and pump operating conditions.

[0075] Safety checks on predictive control quantities can be achieved as follows: Based on the predictive control quantity and the physical model, a second operating parameter of the heating system is determined. When the second operating parameter is within a safe range, the predictive control quantity is determined to be a qualified control quantity, and control commands are generated based on the predictive control quantity. The second operating parameter can be obtained from historical data and physical models of different heating systems. The predictive control quantity of the heating system is mapped to the second operating parameter through the physical model of the heating system. The safe range of the second operating parameter is determined according to industry standards for heating systems.

[0076] In this embodiment, by acquiring the first operating parameters and historical heat load of the first device, the heat load is predicted and the device's health index is calculated. Dynamic constraints are then generated based on the health index, and under these constraints, control quantities to meet the heat load requirements are predicted. Finally, control commands are generated to control the system. This achieves deep coupling between the heating system's operation control and the device's health status. By extracting features from the first operating parameters and mapping them to obtain the health index, the performance status of the devices in the heating system can be quantified and perceived in real time. Constraints for the control quantities are then dynamically generated based on this health index, ensuring that the generated predicted control quantities always remain within the device's current allowable operating range. This mechanism ensures that control commands can adaptively match the actual health status of the devices, thereby effectively slowing down the wear and tear process of devices in a performance degradation state while guaranteeing heating demand, extending the service life of the first device. It also improves the global optimization capability of the heating system under complex operating conditions. By introducing predicted heat loads over a predetermined timeframe based on historical load forecasts, the predicted heat load represents the heating demand that the heating system needs to meet. These predicted heat loads, along with equipment health indices, are used as the basis for calculating predictive control variables. This allows for the determination of the optimal control variables while simultaneously meeting future heat demand and equipment health constraints. This combination of feedforward prediction and health constraints enables the control system to adjust in advance before load fluctuations occur and ensures the health of the equipment, thereby achieving a synergistic improvement in both heating quality assurance and the overall operational efficiency of the heating system.

[0077] The following will describe an exemplary application of the control method of the heating system according to the embodiments of this application in a practical application scenario.

[0078] The control objectives of a heating system typically include: meeting user-side heating demands (e.g., stable room temperature / supply and return water temperatures / pressure), reducing energy consumption (e.g., boiler fuel and pump power consumption), and ensuring equipment safety. In actual operation, the control objectives and equipment performance often present the following contradictions: load fluctuations and thermal inertia: changes in outdoor temperature and user-side heating behavior cause load fluctuations, requiring prediction and advance adjustment; equipment performance differences: circulating pumps, boilers, heat exchangers, valves, and other equipment exhibit varying degrees of aging, scaling, wear, or potential malfunctions; a disconnect between control and fault prediction and Prognostics Health Management (PHM); and safety risks: if aggressive control measures are implemented to maximize load, it may lead to pressure exceeding limits, excessive temperature, pump cavitation, and boiler thermal shock.

[0079] In related technologies, to address the aforementioned contradictions in heating systems, load forecasting-based heating system control has been implemented. This includes combining feedforward neural networks with proportional-integral-derivative (PID) control or model predictive control (MPC), targeting temperature, pressure deviations, and energy consumption. However, these solutions neglect the load-bearing capacity of degraded equipment. For example, a circulating pump with worn bearings and increased vibration may still be forced to operate at high frequencies, accelerating its failure. Furthermore, the heating system cannot proactively reduce rated operating parameters (e.g., voltage, current, power, temperature, frequency) or switch to backup equipment, potentially leading to sudden shutdowns. Equipment efficiency declines with equipment health, and the controller's failure to adjust output control quantities in a timely manner results in increased energy consumption in the heating system.

[0080] Related technologies also combine PHM (Prognostics and Health Management) systems with manual adjustment controls; that is, maintenance personnel manually modify the control parameters of the heating system after receiving an error from the PHM system. However, this method is slow to respond and relies on human experience. Manual adjustments by maintenance personnel are delayed and it is difficult to achieve optimal allocation within minutes. The adjustments may not meet global energy consumption and comfort constraints and are not reproducible. The error rate of manual execution is high, and improper parameter settings may still cause the pressure / temperature of the heating system to exceed limits.

[0081] In this embodiment, addressing the aforementioned problems in related technologies, a control method for a heating system is provided. This method uses short-term heat load forecasts and equipment health indices as inputs to generate a control strategy. When heating equipment is in a performance degradation state, it automatically executes derating operation, load allocation adjustment, or standby switching. Before the control command is output to the Programmable Logic Controller (PLC) or actuator, a physical model-based safety boundary diagnosis and interception mechanism is introduced to clamp or replace commands that may cause operating parameters (e.g., pressure / temperature / flow rate) to exceed limits. Feedback results on the energy consumption and equipment status changes of the heating system after the control command is executed are obtained. Based on these feedback results, the calculation process of the equipment health index is optimized to achieve long-term optimal energy saving and equipment protection.

[0082] For ease of understanding, the control device of the heating system involved in the embodiments of this application will be explained and described. (Reference) Figure 5 , Figure 5 This is a structural diagram of the control device for a heating system provided in an embodiment of this application. The control device for the heating system includes a host computer 501, an edge controller 502, a programmable logic controller, a control cabinet 503, sensors, and actuators 504.

[0083] A Programmable Logic Controller (PLC) is used for low-level I / O acquisition and execution output. The edge controller, which can be an industrial computer (ARM / x86), stores artificial intelligence models and is used to predict health indicators, forecast heating system load, acquire control variables, and perform safety boundary detection on the control flow. Sensors and actuators include temperature sensors, pressure sensors, flow meters, pump / fan frequency converters, regulating valves, and boiler combustion controllers. The host computer can be a Supervisory Control and Data Acquisition (SCADA) system for parameter configuration, curve display, alarms, and historical data archiving. Each module of the heating system's control unit communicates via industrial Ethernet, supporting Network Time Protocol (NTP) synchronization to ensure data time consistency.

[0084] See Figure 4 , Figure 4 This is a flowchart illustrating the control method of the heating system provided in the embodiments of this application, which will be combined with... Figure 4 The steps shown are explained. Figure 4 The entity responsible for executing the steps is Figure 1 Server 200.

[0085] In step 401, data acquisition takes place.

[0086] For example, sensors and actuators collect operating parameters of each device in the heating system and transmit these parameters to the programmable logic controller (PLC). The edge controller reads from the PLC: supply and return water temperatures, network pressure, pump frequency / current, valve opening, outdoor temperature, and zone heat output, etc. Pump frequency typically refers to the power supply frequency of the motor driving the pump.

[0087] In this embodiment, operating parameters of key equipment in the heating system (the first equipment mentioned above) are collected. Key equipment is equipment whose failure would directly lead to heating interruption, significant safety risks, or large-scale failure to meet room temperature standards. Types of key equipment include pumps, heat source equipment, heat exchangers, and valves. Non-key equipment refers to equipment in the heating system that does not affect system energy consumption and operating boundaries, such as auxiliary lighting and valves on terminal branches.

[0088] Pumps include: circulating pumps, pipeline pumps, primary network pumps, and secondary network pumps (including motors and bearings). Pumps directly determine flow rate and pressure difference, and are subject to risks such as cavitation, overload, and vibration degradation.

[0089] Heat source equipment includes: boilers, heat source units, and heat pump units. The heat source equipment determines the heating output and water supply temperature, as well as the risks of thermal shock, overheating, and overpressure.

[0090] Heat exchangers include plate heat exchangers (in-station heat exchange units), which affect heat exchange capacity and secondary network temperature supply.

[0091] Valves include: critical control valves, zone valves, and main valves. Valves affect the main flow distribution and pressure differential.

[0092] In step 402, data quality is checked.

[0093] For example, sensors installed in a heating system may drift or fail, therefore the collected operating parameters need to be quality checked. Data quality checking includes at least one of the following processing steps for the collected operating parameters: outlier removal, redundancy removal, missing value completion, and time alignment.

[0094] In step 403, the health index is obtained.

[0095] For example, the Health Index (HI) is a quantitative indicator used to assess the current health status of equipment or systems. It involves collecting at least one of the following parameters from each key piece of equipment in the heating system: pump or motor vibration, current, and temperature rise; then using a health index prediction model to calculate the health index. The health index prediction model includes neural network models and normalization layers, as well as lightweight machine learning models (e.g., random forests, support vector machines (SVMs), or lightweight convolutional neural networks). The following explanation will illustrate the process of obtaining the health index in conjunction with the hierarchical structure of the health index prediction model.

[0096] For each key piece of equipment, feature extraction is performed using a neural network model based on the vibration, current, and temperature rise values ​​of the pump or motor, resulting in a multi-dimensional feature vector. Specifically, the effective value (RMS), peak value, and kurtosis of the vibration signal are calculated to reflect the total vibration energy level and impact characteristics, obtaining time-domain features based on the vibration signal. Fast Fourier Transform (FFT) or envelope demodulation analysis is performed on the vibration signal to extract the amplitude energy at key frequency points, obtaining frequency-domain features. Motion-Modular Analysis (MCSA) is performed on the current signal to obtain electrical features. The temperature rise rate and steady-state temperature difference are extracted to form thermodynamic features. The thermodynamic, electrical, time-domain, and frequency-domain features are then concatenated into a multi-dimensional feature vector.

[0097] Based on multi-dimensional feature vectors, a lightweight machine learning model (the second prediction model mentioned above) is invoked for prediction processing to obtain the predicted probability that the heating system is in a healthy state. Based on the vibration, current, and temperature rise values ​​of the pump or motor, the normalization function of the normalization layer is invoked. Preset physical limit standards (such as ISO 10816 vibration standard and temperature limits for motor insulation classes) are compared with the collected operating parameters. If the operating parameters are within the standard range, they are mapped to normalized values ​​according to the normalization formula. If the operating parameters are not within the standard range, they are mapped to 0. The predicted probability and the normalized value are weighted and summed to obtain the health index of the key equipment.

[0098] In step 404, dynamic constraints are generated based on the health index.

[0099] For example, dynamic capability constraints are generated for each critical device. The dynamic capability constraint for the maximum pump frequency or maximum flow rate is expressed as follows (1.1): (1.1) The dynamic capability constraint for the maximum rate of ascent is expressed by the following formula (1.2): (1.2) Among them, the lower the health index, the smaller the rate of increase. The rate of increase (Maximum Ramp Rate), also known as the rate of change or oscillation rate, refers to the magnitude of change that a control system allows equipment parameters to undergo per unit time. When... When the value falls below a threshold, a candidate set for standby machine switching is generated (e.g., enabling a standby pump). It is the equipment number (number) (e.g., pumps / boilers / valves) It is a moment equipment The maximum permissible control quantity (such as maximum pump frequency / maximum opening / maximum output). It is the equipment's rated allowable control quantity (rated pump frequency / rated output, etc.); It is a mapping function from health index to traffic (a common example of monotonically non-increasing health index is that the lower the allowable upper limit). It is the change (increase or decrease) of the control quantity in adjacent control cycles. It is a mapping function from the health index to the upper limit of the rate of increase (the lower the health index, the slower the change).

[0100] Step 405 is performed after step 402, in which the heating system load is predicted.

[0101] For example, based on historical heat load and predicted outdoor temperature, the load forecasting model (the first forecasting model mentioned above) is invoked to predict the heat load demand within a future window (e.g., 1 to 6 hours). The load forecasting model can be a combination of a Long Short-Term Memory (LSTM) network and a Gradient Boosting Decision Tree (GBDT) model. Among these, This indicates the current time (either discrete sampling time or continuous time). This refers to the number of prediction steps or the length of the prediction time domain, such as predicting the next 1 to 6 hours. Each step length. () represents the range of a time interval. Used to indicate from time... arrive The predicted heat load sequence. This is the actual heat load.

[0102] After steps 405 and 404, step 406 is executed to obtain the predictive control quantity that satisfies the constraints.

[0103] For example, the types of control variables for a heating system include: pump frequency, valve opening, boiler output (usually referring to the effective heat energy output by the boiler per unit time), and water supply temperature. Based on the predicted heat load, the total rated energy consumption and rated water supply temperature are calculated. At the same time, energy consumption and deviation are minimized, and equipment wear is penalized when the health index decreases. The health index and objective function are expressed as the following formula (2). Meanwhile, the control quantity obtained based on the objective function satisfies the dynamic constraints mentioned above: (2) Where J is the comprehensive objective function, and the optimization objective is to minimize the value of the comprehensive objective function; These are weighting coefficients (used to balance energy consumption, comfort / tracking error, and wear / health constraints); This refers to energy consumption, which can be specifically obtained for each piece of equipment through its respective decision variables. For example, boiler fuel consumption can be obtained based on boiler output, pump power consumption can be obtained based on pump frequency, and valve energy consumption can be obtained based on valve opening degree. It is the total fixed energy consumption, calculated based on the predicted heat load; It is the actual water supply temperature value, which can be the primary network or the secondary network water supply temperature, and is determined by predicting the heat load. It sets the water supply temperature value or target temperature trajectory; It is the square norm; It is a wear / deterioration penalty function, which increases with decreasing health index (HI decreases) and increasing load / control quantity ( (Increases) and increases. . It is a control variable (which can be a vector such as pump frequency, valve opening, boiler output, etc.). It is the weight value; The function's value increases as the health index decreases and the load increases. The comprehensive objective function is used to calculate the predictive control variable that minimizes the objective function for each device. The comprehensive objective function can be solved using Model Predictive Control (MPC) to obtain the predictive control variable for each critical device.

[0104] In step 407, a safety check is performed on the predictive control quantity.

[0105] For example, a physical model is a simplified expression that uses mathematical formulas and logical rules to describe a real-world physical object or process. Setting up a physical model involves safety checks on control quantities predicted by artificial intelligence to prevent these predicted control quantities from exceeding corresponding safety boundaries. After receiving a control command, the heating system executes the corresponding control command and generates corresponding operating parameters. For example, the control quantities carried by the control command include: pump frequency, valve opening, boiler output, and water supply temperature. The generated operating parameters include: pressure, temperature, flow rate, and pump operating conditions.

[0106] Safety checks on predictive control quantities can be achieved as follows: Based on the predictive control quantities and the physical model, the predictive operating parameters of the heating system (the second operating parameter mentioned above) are determined. When the predictive operating parameters are within the safe range, the predictive control quantity is determined to be a qualified control quantity, and control commands are generated based on the predictive control quantity. The predictive operating parameters can be obtained from historical data and physical models of different heating systems. The predictive control quantities of the heating system are mapped to the predictive operating parameters through the physical model of the heating system. The safe range of the predictive operating parameters is determined according to industry standards for heating systems.

[0107] Among the controlled variables, pump frequency directly affects flow rate and pressure; valve opening directly affects pipeline resistance and differential pressure; and boiler output directly affects the actual water supply temperature. Determine whether the above operating parameters are within safe ranges, and based on the determination results, determine whether the controlled variables are qualified. Pressure types include: supply main pressure, return main pressure, pump suction (inlet) pressure, and pump discharge (outlet) pressure; the difference between supply main pressure and return main pressure is the supply and return pressure differential. The difference between pump suction pressure and pump discharge pressure is the pump differential pressure.

[0108] For example, taking flow rate as an example, the flow rate is obtained based on pump frequency and valve opening, and is expressed as the following formula (3): (3) Where H is the pump head, approximately equal to... That is, the product of the preset coefficient and the pump frequency score. S is the total resistance coefficient, determined based on the valve opening O, and approximated by... , This is the resistance coefficient of a fixed portion of the pipeline network. Similarly, the predictive operating parameters corresponding to the control variables are calculated using a physical model.

[0109] In some embodiments, if the predictive operating parameters corresponding to the predictive control quantity exceed the safety range, the predictive control quantity is determined to be unqualified, and is discarded and reacquired. For unqualified predictive control quantities, alarms and logs are generated. The logs record the triggered constraints, measurement points / estimated values, candidate control quantities and the final issued control quantity, and the health indicators of the associated devices controlled by the predictive control quantity. Logs were used as training sample pairs to incrementally train the MPC model for control variables, the health index prediction model, and the load prediction model.

[0110] In some embodiments, if the predicted operating parameter corresponding to the predicted control quantity exceeds the safe range, the predicted control quantity is determined to be unqualified, a clamping control quantity is generated, and a control command is generated based on the clamping control quantity. Clamping refers to the process of holding a parameter at a reference value. This reference value can be fixed or adjustable. The clamping control quantity is preset to keep the predicted operating parameter within the safe range.

[0111] In some embodiments, if the predicted control quantity continues to be unqualified, it indicates that the equipment in the heating system itself may be faulty, and a pump set switch or a new pump set or heating boiler should be performed.

[0112] In some embodiments, the MPC model, health index prediction model, and load prediction model may fail, for example, due to a failure of the model itself or a failure of the server running the model. When the model is unavailable, communication with the server running the model is interrupted, and control quantities are generated using rule control or PID control to ensure continuous heating.

[0113] In step 408, control commands are sent to the heating system, and the operating data of the heating system is monitored in real time.

[0114] For example, verified control commands are written into frequency converters and valves via programmable logic controllers; the execution response and deviation of each key device in the heating system are monitored, and if execution fails or deviation continues to increase, a degradation strategy or manual intervention is triggered.

[0115] In step 409, the feedback results of the heating system are obtained and stored.

[0116] For example, actual energy consumption, room temperature compliance rate, and changes in equipment health are written into a historical database. The data in the historical database can be used for load forecasting model recalibration and health index forecasting model drift correction.

[0117] In this embodiment, load forecasting reduces excessive heating and frequent adjustments, and higher-efficiency equipment is selected to handle the load under health index constraints, thereby improving overall energy efficiency and reducing the energy consumption of the heating system. When equipment performance in the heating system deteriorates, automatic derating and switching to backup units reduces the probability of equipment failure escalating into a system-wide shutdown. Physical safety boundary checks intercept unreasonable control quantities, preventing erroneous control outputs from artificial intelligence predictions that could lead to equipment damage or pipeline accidents.

[0118] The following description continues to illustrate the exemplary structure of the control device 455 for the heating system provided in this application embodiment as a software module. In some embodiments, such as... Figure 2 As shown, the software module in the control device 455 of the heating system stored in the memory 450 may include: The data acquisition module 4551 is used to acquire the first operating parameters of each first device in the heating system and the historical heat load; the load prediction module 4552 is used to predict the predicted heat load of the heating system based on the historical heat load, wherein the predicted heat load is the heat load of the heating system within a preset time period in the future; the index prediction module 4553 is used to extract features from the first operating parameters of each first device to obtain a first feature vector of each first device; to perform mapping processing on the first feature vector of each first device to obtain a health index of each first device, wherein the health index is a parameter used to measure the performance status of the first device; to generate constraints on the control quantity of the heating system based on the health index; the control module 4554 is used to predict the predicted control quantity that satisfies the constraints based on the health index and the predicted heat load; to generate control commands based on the predicted control quantities; and to control the heating system based on the control commands.

[0119] In some embodiments, the load prediction module 4552 is configured to: acquire the predicted outdoor temperature within the preset time period; perform feature extraction processing on a first prediction model based on the historical heat load to obtain a first feature sequence, wherein the first feature sequence includes a second feature vector corresponding to the load at different times; perform feature extraction processing on the first prediction model based on the predicted outdoor temperature within the preset time period to obtain a second feature sequence, wherein the second feature sequence includes a temperature feature vector corresponding to the temperature at different times; perform iterative prediction processing on the first prediction model based on the first feature sequence and the second feature sequence to obtain multiple sub-predicted heat loads at different times, wherein the output of each iteration of the iterative prediction processing is the sub-predicted heat load at different times, and the input of each iteration of the iterative prediction processing is the already predicted sub-predicted heat load, the first feature sequence, and the second feature sequence; and combine each of the sub-predicted heat loads into a predicted heat load in sequence form.

[0120] In some embodiments, the index prediction module 4553 is configured to perform the following processing for each of the first devices: extract features for each of the first operating parameters of the first device to obtain multiple parameter features; and concatenate each parameter feature to obtain the first feature vector.

[0121] In some embodiments, the index prediction module 4553 is configured to perform the following processing for each of the first devices: calling a second prediction model based on the first feature vector to predict the probability that the first device is in a healthy state; normalizing the first feature vector to obtain a normalized value; and weighted summing the predicted probability and the normalized value to obtain the health index.

[0122] In some embodiments, the control module 4554 is configured to calculate actual energy consumption parameters based on the predicted heat load; construct an objective function based on the actual energy consumption parameters, the health index, and the control variables; call the predictive control model to calculate the target value of the control variable that minimizes the value of the objective function and satisfies the constraint conditions, and use the target value as the predictive control quantity.

[0123] In some embodiments, the actual energy consumption parameters include: total fixed energy consumption and actual water supply temperature; the control variables include: set water supply temperature and equipment control parameters; the control module 4554 is used to calculate the norm of the square of the difference between the actual water supply temperature and the set water supply temperature; calculate a first ratio between the sum of the equipment energy consumption of each of the first devices and the total fixed energy consumption; calculate a second ratio between the equipment control parameters of each of the first devices and the health index; perform a weighted summation on each of the second ratios to obtain a first weighted result; and use the weighted summation of the first weighted result, the first ratio, and the norm of the square as the objective function.

[0124] In some embodiments, the control module 4554 is configured to perform calculations based on the predicted control quantity and the physical model of the heating system to obtain a second operating parameter of the heating system; and to perform the following processing for each second operating parameter: detecting whether the second operating parameter is within a safe range; in response to the second operating parameter being within a safe range, determining the predicted control quantity associated with the second operating parameter; and generating a control command corresponding to the associated predicted control quantity.

[0125] This application provides a computer program product, which includes a computer program or computer-executable instructions. A processor executes the computer program or computer-executable instructions to implement the heating system control method described above in this application.

[0126] This application provides a computer-readable storage medium storing computer-executable instructions or a computer program. When the computer-executable instructions or the computer program are executed by a processor, the processor will execute the control method of the heating system provided in this application. For example, ... Figure 3A The control method of the heating system is shown.

[0127] In some embodiments, the computer-readable storage medium may be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface memory, optical disk, or CD-ROM; or it may be a variety of devices including one or any combination of the above-mentioned memories.

[0128] In some embodiments, computer-executable instructions may take the form of programs, software, software modules, scripts, or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as stand-alone programs or as modules, components, subroutines, or other units suitable for use in a computing environment.

[0129] As an example, computer-executable instructions may, but do not necessarily, correspond to files in a file system. They may be stored as part of a file that holds other programs or data, for example, in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple co-located files (e.g., a file that stores one or more modules, subroutines, or code sections).

[0130] As an example, computer-executable instructions can be deployed to execute on a single electronic device, or on multiple electronic devices located at one location, or on multiple electronic devices distributed across multiple locations and interconnected via a communication network.

[0131] In summary, through the embodiments of this application, by acquiring the first operating parameters and historical heat load of the first device, predicting the heat load and calculating the device's health index based on these parameters, dynamic constraints are generated based on the health index, and control quantities to meet the heat load requirements are predicted under these constraints. Finally, control commands are generated to control the system. This achieves deep coupling between the operation control of the heating system and the health status of the equipment. By extracting features from the first operating parameters and mapping them to obtain the health index, the performance status of the equipment in the heating system can be quantified and perceived in real time. The constraints for dynamically generating control quantities based on this health index ensure that the generated predicted control quantities always remain within the current allowable operating range of the equipment. This mechanism ensures that the control commands can adaptively match the actual health status of the equipment, thereby effectively slowing down the wear and tear process of equipment in a performance degradation state while ensuring heating demand, and extending the service life of the first device. It also improves the global optimization capability of the heating system under complex operating conditions. By introducing predicted heat loads over a predetermined timeframe based on historical load forecasts, the predicted heat load represents the heating demand that the heating system needs to meet. These predicted heat loads, along with equipment health indices, are used as the basis for calculating predictive control variables. This allows for the determination of the optimal control variables while simultaneously meeting future heat demand and equipment health constraints. This combination of feedforward prediction and health constraints enables the control system to adjust in advance before load fluctuations occur and ensures the health of the equipment, thereby achieving a synergistic improvement in both heating quality assurance and the overall operational efficiency of the heating system.

[0132] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, and improvements made within the spirit and scope of this application are included within the scope of protection of this application.

Claims

1. A control method for a heating system, characterized in that, The method includes: Obtain the first operating parameters and historical heat load of each primary device in the heating system; Based on the historical heat load, the predicted heat load of the heating system is predicted, wherein the predicted heat load is the heat load of the heating system within a preset time period in the future; Feature extraction is performed on the first operating parameters of each first device to obtain a first feature vector for each first device; Based on the first feature vector of each of the first devices, a mapping process is performed to obtain the health index of each of the first devices, wherein the health index is a parameter used to measure the performance status of the first devices; Constraints for generating control quantities of the heating system based on the health index; Based on the health index and the predicted heat load, predictive control quantities that satisfy the constraints are predicted. Control commands are generated based on the predicted control values, and the heating system is controlled based on the control commands.

2. The method according to claim 1, characterized in that, The prediction of the predicted heat load of the heating system based on the historical heat load includes: Obtain the predicted outdoor temperature within the preset time period; Based on the historical heat load, the first prediction model is called to perform feature extraction processing to obtain a first feature sequence, wherein the first feature sequence includes a second feature vector corresponding to the load at different times; Based on the predicted outdoor temperature within the preset time period, the first prediction model is invoked to perform feature extraction processing to obtain a second feature sequence, wherein the second feature sequence includes temperature feature vectors corresponding to temperatures at different times. Based on the first feature sequence and the second feature sequence, the first prediction model is invoked to perform iterative prediction processing to obtain multiple sub-predicted heat loads at different times. The output of each iteration of the iterative prediction processing is the sub-predicted heat load at different times, and the input of each iteration of the iterative prediction processing is the already predicted sub-predicted heat load, the first feature sequence, and the second feature sequence. Each of the sub-predicted heat loads is combined into a sequence of predicted heat loads.

3. The method according to claim 1, characterized in that, The step of extracting features from the first operating parameters of each first device to obtain a first feature vector for each first device includes: For each of the first devices, the following processing is performed: For each of the first operating parameters of the first device, feature extraction is performed to obtain multiple parameter features; Each of the parameter features is concatenated to obtain the first feature vector.

4. The method according to claim 1, characterized in that, The mapping process based on the first feature vector of each of the first devices to obtain the health index of each of the first devices includes: For each of the first devices, the following processing is performed: The second prediction model is invoked based on the first feature vector to predict the probability that the first device is in a healthy state. The first feature vector is normalized to obtain a normalized value; The health index is obtained by weighted summation of the predicted probability and the normalized value.

5. The method according to claim 1, characterized in that, The prediction of the predictive control quantity that satisfies the constraints based on the health index and the predicted heat load includes: Calculate the actual energy consumption parameters based on the predicted heat load; A target function is constructed based on the actual energy consumption parameters, the health index, and the control variables. The predictive control model is invoked to calculate the target value of the control variable that minimizes the value of the objective function and satisfies the constraints, and the target value is used as the predictive control variable.

6. The method according to claim 5, characterized in that, The actual energy consumption parameters include: total fixed energy consumption and actual water supply temperature; the control variables include: set water supply temperature and equipment control parameters. The construction of the objective function based on the actual energy consumption parameters, the health index, and control variables includes: Calculate the square norm of the difference between the actual water supply temperature value and the set water supply temperature value; Calculate a first ratio between the sum of the energy consumption of each of the first devices and the total fixed energy consumption; Calculate a second ratio between the device control parameter and the health index for each of the first devices, and perform a weighted summation on each second ratio to obtain a first weighted result; The weighted sum of the first weighted result, the first ratio, and the square norm is used as the objective function.

7. The method according to claim 1, characterized in that, The generation of control commands based on the predicted control quantity includes: The second operating parameters of the heating system are obtained by performing calculations based on the predicted control quantity and the physical model of the heating system. For each of the second working parameters, perform the following processing: Detect whether the second operating parameter is within a safe range, and in response to the second operating parameter being within a safe range, determine the predictive control quantity associated with the second operating parameter; Generate the control command corresponding to the associated predictive control quantity.

8. A control device for a heating system, characterized in that, The device includes: The data acquisition module is used to acquire the first operating parameters of each primary device in the heating system, as well as the historical heat load; The load prediction module is used to predict the predicted heat load of the heating system based on the historical heat load, wherein the predicted heat load is the heat load of the heating system within a preset time period in the future. An index prediction module is used to extract features from the first operating parameters of each of the first devices to obtain a first feature vector for each of the first devices; perform mapping processing on the first feature vector of each of the first devices to obtain a health index for each of the first devices, wherein the health index is a parameter used to measure the performance status of the first devices; and generate control constraints for the heating system based on the health index. The control module is used to predict the predictive control quantity that satisfies the constraints based on the health index and the predicted heat load; generate control commands based on the predictive control quantity; and control the heating system based on the control commands.

9. An electronic device, characterized in that, The electronic device includes: Memory is used to store executable instructions or computer programs. A processor, when executing computer-executable instructions or computer programs stored in the memory, implements the control method of the heating system according to any one of claims 1 to 7.

10. A computer-readable storage medium storing computer-executable instructions or a computer program, characterized in that, When the computer-executable instructions or computer program are executed by a processor, the control method of the heating system according to any one of claims 1 to 7 is implemented.

Images