Energy efficiency control method for low-temperature waste heat recovery heating system

Abstract

The present application relates to the field of industrial waste heat recovery and heating control, and discloses an energy efficiency control method of a low-temperature waste heat recovery heating system, which collects heat source side parameters, condenser side parameters, compressor operation parameters, environmental parameters and real-time electricity prices to form system operation data covering heat sources, heat pump units, heating loads and economic signals, thereby providing a complete data basis for energy efficiency control. The model parameters of the equivalent thermal resistance-thermal capacity model are identified online by using the recursive least squares algorithm, the model can be continuously corrected according to the actual operation state of the system, the dependence on the pre-accurately obtained physical parameters of the equipment and the pipe network is reduced, and the adaptability of the control method to the heat source fluctuation, the equipment performance change and the unmodeled dynamics is improved. The prediction time domain of the model predictive control is dynamically adjusted according to the change rate of the available waste heat power of the heat source and the heating load prediction error, so that the control response speed can be improved when the heat source changes rapidly, and the operation optimization of a long period can be considered under stable working conditions.

Description

Technical Field

[0001] This invention relates to the field of industrial waste heat recovery and heating control, specifically to an energy efficiency control method for a low-temperature waste heat recovery heating system. Background Technology

[0002] Low-temperature waste heat is widely present in high-energy-consuming industrial production processes such as steel, chemical, and cement industries. It is typically discharged in the form of low-temperature circulating water, process cooling water, and geothermal tailwater, with temperatures mostly ranging from 30℃ to 65℃. Due to the low grade of this type of waste heat, it is difficult to use it directly for conventional heating. Large amounts of this low-temperature waste heat are directly released into the external environment, causing not only energy waste but also potential thermal pollution. With the development of heat pump technology and heat exchange network technology, upgrading low-temperature waste heat through heat pump units and then delivering the upgraded heat to heating users has become an important way to improve industrial energy efficiency.

[0003] Existing low-temperature waste heat recovery heating systems typically employ fixed temperature difference, constant flow rate, or proportional-integral-derivative (PID) control methods, adjusting the compressor and circulating pump operation based on the temperature deviation between the supply and return water. However, the temperature and flow rate of waste heat sources during industrial production are easily affected by factors such as production load, equipment start-up and shutdown, and process switching, exhibiting strong randomness and time-varying characteristics. Simultaneously, the user-side heating load also changes with outdoor temperature, building thermal inertia, and heating demand. Traditional control methods struggle to comprehensively consider heat source fluctuations and heat load variations, resulting in relatively delayed control responses. This can easily lead to problems such as supply water temperature fluctuations, insufficient heating, or frequent start-ups and shutdowns of heat pump units when heat source power decreases rapidly.

[0004] Existing energy efficiency control methods typically focus on improving the instantaneous energy efficiency ratio (EER) of heat pump units, with limited consideration given to equipment health factors such as compressor exhaust temperature, operating frequency variations, and surge risk. They also fail to fully utilize time-of-use electricity pricing signals for economical dispatching. Simply pursuing a high instantaneous EER may lead to prolonged operation of the compressor under high thermal stress or unstable conditions, resulting in reduced equipment lifespan. Maintaining high heat pump output during periods of high electricity prices also increases the system's overall operating costs. Furthermore, in systems equipped with thermal storage devices, existing control methods struggle to coordinate the dispatching of the heat pump and thermal storage devices based on the residual heat margin of the heat source, the state of thermal storage, and peak-valley electricity price fluctuations.

[0005] Meanwhile, existing model predictive control methods typically rely on relatively accurate thermodynamic models, requiring prior acquisition of physical parameters such as thermal resistance and heat capacity of heat exchangers, heating networks, and buildings, as well as the deployment of numerous measuring devices. Due to the complex network structure of actual systems and the changing performance of equipment over time, accurate determination of model parameters is difficult. Furthermore, a fixed prediction time domain cannot simultaneously consider control response capabilities during rapid heat source fluctuations and global economic efficiency during steady-state operation, thus limiting the application of model predictive control methods in existing waste heat heating systems. Summary of the Invention

[0006] The purpose of this invention is to overcome the problems of existing control methods for low-temperature waste heat recovery heating systems, such as high dependence on accurate physical models and poor adaptability to heat source and load fluctuations, and to provide an energy efficiency control method for low-temperature waste heat recovery heating systems.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides an energy efficiency control method for a low-temperature waste heat recovery heating system, comprising the following steps: Collect parameters from the heat source side, condenser side, compressor operating parameters, environmental parameters, and real-time electricity price to obtain system operating data; Based on system operation data, the model parameters of the equivalent thermal resistance-heat capacity model are identified online using the recursive least squares algorithm. The dynamic prediction time domain of the model predictive control is determined based on the rate of change of available waste heat power of the heat source and the heat load prediction error. The three-dimensional comprehensive efficiency index is determined based on the instantaneous energy efficiency ratio, the compressor thermal stress coefficient, and the time-of-use electricity price penalty factor. Based on the equivalent thermal resistance-heat capacity model, dynamic prediction time domain and three-dimensional comprehensive efficiency index, a multi-objective constrained optimization model is established for solution to obtain the optimal control sequence of compressor speed, heat source side circulating pump frequency and secondary network circulating pump frequency; Execute the first control vector in the optimal control sequence, update the system state based on the deviation between the actual output value and the model prediction value, and perform rolling optimization; The protection decision is made based on the compressor thermal stress coefficient, evaporator outlet temperature and heat pump suction-discharge pressure ratio. When the preset protection conditions are met, the operating parameters of the compressor and the heat source side circulation pump are adjusted.

[0008] A further improvement of the present invention is that the heat source side parameters include evaporator inlet temperature, evaporator outlet temperature, heat source flow rate, and heat source temperature change rate. Condenser-side parameters include condenser inlet temperature, condenser outlet temperature, and secondary network circulation flow rate; Compressor operating parameters include compressor speed, input power, discharge temperature, suction temperature, and suction-discharge pressure ratio; Environmental parameters include outdoor temperature and wind speed; The collected data is processed by removing spike noise, low-pass filtering, and repairing missing data to obtain system operation data.

[0009] A further improvement of the present invention is that the equivalent thermal resistance-heat capacity model includes a first-order equivalent thermal resistance-heat capacity link or a second-order equivalent thermal resistance-heat capacity link that respectively characterizes the thermal dynamic characteristics of the heat source side, the heat exchanger, the heating network and the heated building. The model parameters are updated using a recursive least squares algorithm with a forgetting factor. When the rate of change of the heat source temperature exceeds 2℃ / min, the forgetting factor is reduced to below 0.92. When the heat source temperature stabilizes, the forgetting factor is restored to above 0.98.

[0010] A further improvement of this invention is that the dynamic prediction time domain is determined according to the following formula:

[0011] In the formula, For the dynamic prediction time domain at the current moment, For the time domain of prediction, The rate of change of available waste heat power from the heat source. The nominal waste heat power, For heat load prediction error, For nominal heat load, and For adjustment coefficients, and These are the lower and upper bounds of the prediction time domain, respectively; the lower bound is 15 minutes and the upper bound is 120 minutes.

[0012] A further improvement of this invention lies in determining the three-dimensional comprehensive performance index according to the following formula:

[0013] In the formula, It is a three-dimensional comprehensive efficiency index. Instantaneous energy efficiency ratio, To preset the maximum energy efficiency ratio, The thermal stress coefficient of the compressor, As a time-of-use electricity pricing penalty factor, , and These are energy efficiency weight, thermal stress weight, and electricity price weight, respectively. The weights are adjusted based on the waste heat margin of the heat source, the compressor exhaust temperature, and the real-time electricity price.

[0014] A further improvement of this invention is that the optimization objectives of the multi-objective constrained optimization model include water supply temperature tracking error, compressor speed change penalty, and low three-dimensional comprehensive efficiency index penalty; An improved particle swarm optimization algorithm is used to solve the multi-objective constrained optimization model. The improved particle swarm optimization algorithm adopts an inertial weight that decays with the number of iterations, sets an early stopping threshold and a chaotic mutation operator, and applies chaotic mapping perturbation to some particles when the global optimum has not been updated in several consecutive iterations.

[0015] A further improvement of this invention is that, based on the deviation between the condenser outlet temperature actually measured at the current moment and the predicted temperature of the equivalent thermal resistance-heat capacity model, the subsequent predicted temperature in the dynamic prediction time domain is uniformly corrected; and the state variables of the equivalent thermal resistance-heat capacity model are updated using the Kalman filter algorithm according to a preset update cycle, which is 30 minutes.

[0016] A further improvement of the present invention is that when the thermal stress coefficient of the compressor exceeds 0.85 and the duration reaches 5 minutes, the compressor acceleration rate is limited to no more than 0.5 Hz / s, and the compressor frequency is reduced according to a preset frequency reduction cycle until the thermal stress coefficient of the compressor is reduced to below 0.6. When the evaporator outlet temperature is below 32℃ and the heat pump suction-discharge pressure ratio exceeds 5.5, the compressor frequency is reduced at a rate of 0.2Hz / s, and the frequency of the heat source side circulation pump is increased until the preset danger zone is removed or an alarm is triggered.

[0017] A further improvement of this invention is that it also includes an auxiliary scheduling step for the thermal storage tank, which determines the change in the equivalent state of charge of the thermal storage tank according to the following formula:

[0018] In the formula, This represents the equivalent state of charge of the thermal storage tank. and These are the heat charging efficiency and the heat dissipation efficiency, respectively. and These are the charging power and the discharging power, respectively. The equivalent heat capacity of the thermal storage tank is used; the thermal storage tank is controlled to charge during periods of low electricity price and when the heat source has surplus heat; the thermal storage tank is controlled to release heat during periods of peak electricity price; and the equivalent state of charge and heat release rate of the thermal storage tank are incorporated into the multi-objective constrained optimization model.

[0019] Secondly, the present invention provides an energy efficiency control system for a low-temperature waste heat recovery heating system, comprising: The data acquisition module is used to collect parameters from the heat source side, condenser side, compressor operating parameters, environmental parameters, and real-time electricity prices to obtain system operating data. The model identification module is used to identify the model parameters of the equivalent thermal resistance-heat capacity model online based on system operation data using a recursive least squares algorithm. The prediction time domain determination module is used to determine the dynamic prediction time domain of the model predictive control based on the rate of change of available waste heat power of the heat source and the prediction error of the heat load. The efficiency index determination module is used to determine the three-dimensional comprehensive efficiency index based on the instantaneous energy efficiency ratio, compressor thermal stress coefficient, and time-of-use electricity price penalty factor. The optimization solution module is used to solve a multi-objective constrained optimization model based on the equivalent thermal resistance-heat capacity model, dynamic prediction time domain and three-dimensional comprehensive efficiency index, and obtain the optimal control sequence of compressor speed, heat source side circulating pump frequency and secondary network circulating pump frequency. The feedback correction module is used to execute the first control vector in the optimal control sequence, update the system state based on the deviation between the actual output value and the model prediction value, and perform rolling optimization. The protection control module is used to make protection judgments based on the compressor thermal stress coefficient, evaporator outlet temperature and heat pump suction and discharge pressure ratio, and adjust the operating parameters of the compressor and heat source side circulation pump when the preset protection conditions are met.

[0020] Compared with the prior art, the present invention has the following beneficial effects: This invention collects system operation data covering heat sources, heat pump units, heating loads, and economic signals by collecting parameters from the heat source side, condenser side, compressor operating parameters, environmental parameters, and real-time electricity prices, providing a complete data foundation for energy efficiency control. It employs a recursive least squares algorithm to identify the model parameters of the equivalent thermal resistance-heat capacity model online, continuously revising the model based on the actual system operating state. This reduces reliance on pre-acquired, precisely obtained equipment and pipeline physical parameters, improving the control method's adaptability to heat source fluctuations, equipment performance changes, and unmodeled dynamics. The prediction time domain of the model predictive control is dynamically adjusted based on the change rate of available waste heat power from the heat source and the heat load prediction error, improving control response speed during rapid heat source changes and balancing long-term operational optimization under stable conditions. Furthermore, this invention integrates instantaneous energy efficiency ratio, compressor thermal stress coefficient, and time-of-use electricity price penalty factor into a three-dimensional comprehensive efficiency index and establishes a multi-objective constrained optimization model. This model collaboratively determines compressor speed, heat source-side circulating pump frequency, and secondary network circulating pump frequency, avoiding the increase in equipment thermal stress or operating costs caused by solely pursuing instantaneous energy efficiency. By combining the execution of the first control vector, output deviation correction, and system status update, rolling optimization is achieved, improving the accuracy of water supply temperature tracking and operational stability. At the same time, protective control is implemented based on compressor thermal stress, evaporator outlet temperature, and suction-discharge pressure ratio, which can reduce the risk of thermal shock and surge, thus taking into account heating stability, system energy efficiency, equipment safety, and operational economy. Attached Figure Description

[0021] Figure 1 This is a flowchart of the present invention; Figure 2 This is a system diagram of the present invention. Detailed Implementation

[0022] To further understand the content of this invention, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments are merely illustrative and not limiting of the invention.

[0023] See Figure 1 The energy efficiency control method of the low-temperature waste heat recovery heating system of the present invention includes the following steps.

[0024] Step S1: Collect parameters from the heat source side, condenser side, compressor operating parameters, environmental parameters, and real-time electricity price, and process the collected data to obtain system operating data.

[0025] The system collects various operating parameters according to a preset sampling period, which can be determined based on the response speed of the control system and on-site communication conditions. In this embodiment, data acquisition and model parameter updates can be completed in a 30-second cycle.

[0026] The parameters on the heat source side include evaporator inlet temperature, evaporator outlet temperature, heat source flow rate, and heat source temperature change rate. Among them, the evaporator inlet temperature represents the temperature of the low-temperature waste heat medium when it enters the evaporator, the evaporator outlet temperature represents the temperature of the low-temperature waste heat medium when it leaves the evaporator after releasing heat, and the heat source temperature change rate is used to characterize the degree of fluctuation of the waste heat source caused by changes in industrial production conditions.

[0027] The condenser-side parameters include the condenser inlet temperature, the condenser outlet temperature, and the secondary network circulation flow rate. The condenser inlet temperature is the secondary network return water temperature, the condenser outlet temperature is the secondary network supply water temperature, and the secondary network circulation flow rate is used to determine the heat delivered by the condenser to the user side.

[0028] Compressor operating parameters include compressor speed, compressor input power, compressor discharge temperature, compressor suction temperature, and heat pump suction-discharge pressure ratio. Compressor speed can be characterized by the compressor operating frequency; compressor input power is used to calculate the instantaneous energy efficiency ratio of the heat pump unit; discharge temperature and its changes are used to evaluate compressor thermal stress; and suction-discharge pressure ratio is used to identify whether the compressor is approaching the surge danger zone.

[0029] Environmental parameters include outdoor temperature and wind speed. Outdoor temperature characterizes changes in the building's heating load, while wind speed corrects for heat dissipation from the building envelope. Economic parameters include real-time electricity prices published by the power grid or energy management platform, as well as the corresponding peak, flat, and off-peak electricity price periods.

[0030] For the collected data, a spike noise identification method is first used to remove sensor spike data that significantly deviates from the normal range of variation. Then, a low-pass filter is used to suppress high-frequency measurement noise. For short-term missing data, a linear interpolation method is used to complete the data. The interpolation results are then verified in conjunction with thermodynamic constraints such as the supply and return water temperature relationship, flow continuity, and equipment operating status. This yields system operating data that can be used for subsequent model identification and optimized control.

[0031] This embodiment prioritizes reading the existing temperature, flow rate, pressure, power, and equipment operation signals in the original heating system, without requiring additional high-precision heat flux density sensors. For systems with existing flow measurement points, the existing flow signals can be directly accessed, thereby reducing the cost of modifying the existing system.

[0032] Step S2: Based on the system operation data, the recursive least squares algorithm is used to identify the model parameters of the equivalent thermal resistance-heat capacity model online.

[0033] To avoid establishing complex three-dimensional computational fluid dynamics models or thermodynamic models that rely on all the physical parameters of the equipment, this embodiment uses a reduced-order equivalent thermal resistance-heat capacity model to describe the thermal dynamic characteristics of the low-temperature waste heat recovery heating system. Equivalent thermal resistance characterizes the heat transfer resistance between different media, equipment, and buildings, while equivalent heat capacity characterizes the ability of equipment, piping networks, and buildings to store heat.

[0034] Specifically, the heat source side, heat exchanger, heating network, and heated building are respectively equivalent to a first-order equivalent thermal resistance-heat capacity link or a second-order equivalent thermal resistance-heat capacity link. Among them, the equivalent link on the heat source side is used to describe the dynamic relationship between heat source temperature, heat source flow rate, and evaporator outlet temperature; the equivalent link on the heat exchanger side is used to describe the heat exchange process inside the evaporator, condenser, and plate heat exchanger; the equivalent link on the heating network is used to describe the supply and return water temperatures and the thermal inertia of the network; and the equivalent link on the heated building is used to describe the dynamic relationship between outdoor temperature, heating capacity, and indoor average temperature.

[0035] The identification data for the equivalent thermal resistance-heat capacity model may include evaporation temperature, condensation temperature, supply and return water temperature difference, ambient temperature, compressor input power, heat source flow rate, and secondary network circulation flow rate. Model output may include evaporator outlet temperature, condenser outlet temperature, and the average indoor temperature of the heated building.

[0036] A recursive least squares algorithm with a forgetting factor is used to update the model parameters corresponding to the equivalent thermal resistance and equivalent heat capacity online. The recursive least squares algorithm uses the input and output data at the current sampling time as a basis to recursively correct the model parameter estimation results at the previous sampling time. Therefore, it is not necessary to obtain all physical parameters such as pipe material, pipe length, insulation layer thickness, soil thermal properties, building envelope thermal resistance, and internal heat transfer coefficient of equipment in advance, nor is it necessary to stop the operation of the heating system for special model calibration.

[0037] The forgetting factor is adjusted based on the degree of fluctuation in the heat source temperature. When the rate of change of the heat source temperature exceeds 2℃ / min, it indicates that the heat source has entered a rapid change state. The forgetting factor is temporarily reduced to below 0.92 to give newer operating data a higher weight in parameter identification and improve the model parameters' tracking speed of heat source changes. When the heat source temperature stabilizes, the forgetting factor is restored to above 0.98 to reduce the parameter estimation variance and improve the model's noise resistance under stable operating conditions.

[0038] In this embodiment, the model parameters are updated every 30 seconds. Since the established equivalent thermal resistance-thermal capacity model is a reduced-order model, there may be unmodeled dynamic errors. However, these errors can be compensated for in the subsequent model predictive control output deviation feedback correction and Kalman filter state update process. Therefore, there is no need to pursue a highly complex first-principles model.

[0039] Step S3: Determine the dynamic prediction time domain of the model predictive control based on the rate of change of available waste heat power of the heat source and the heat load prediction error.

[0040] The available waste heat power of the heat source can be determined based on the heat source flow rate, evaporator inlet temperature, evaporator outlet temperature, and the specific heat capacity of the heat source medium. The rate of change of the available waste heat power is obtained by differentiating the available waste heat power at adjacent sampling times. The heat load prediction error is the deviation between the actual heating load and the predicted heating load for the corresponding time period.

[0041] The time domain for dynamic prediction is determined by the following formula:

[0042] In the formula, For the dynamic prediction time domain at the current moment; For the baseline prediction time domain, 60 minutes is used in this embodiment; This represents the rate of change of available waste heat power at the current moment. Nominal waste heat power; This is the error in heat load prediction; Nominal heat load; and This is the adjustment coefficient; and These are the lower bound and upper bound of the prediction time domain, respectively.

[0043] In this embodiment, the lower limit of the prediction time domain is 15 minutes, the upper limit of the prediction time domain is 120 minutes, and the control time domain is fixed at 12 minutes. When the available waste heat power of the heat source decays rapidly or the heat load prediction error increases, the exponential term decreases, and the dynamic prediction time domain is automatically shortened to 15-30 minutes. This allows the model predictive control to focus more on the recent operating status, reduces the impact of long-term prediction errors on the current control decision, and avoids overshooting during periods of rapid changes in the heat source.

[0044] When the heat source power and heat load are in a stable state, the change rate of available waste heat power and the prediction error of heat load are both small. The dynamic prediction time domain is close to the base prediction time domain or extended to 90-120 minutes, enabling the control system to comprehensively consider energy efficiency, electricity price and heat storage scheduling over a longer time range, thereby improving the overall economy of the system.

[0045] Step S4: Determine the three-dimensional comprehensive efficiency index based on the instantaneous energy efficiency ratio, compressor thermal stress coefficient, and time-of-use electricity price penalty factor.

[0046] The instantaneous energy efficiency ratio (EER) is the ratio of condenser heat release to compressor input power. Condenser heat release can be calculated based on the secondary network circulation flow rate, the temperature difference between the secondary network supply and return water, and the specific heat capacity of water. The compressor thermal stress coefficient is determined by weighting the normalized cumulative deviation of the compressor discharge temperature relative to the safety upper limit and the normalized value of the discharge temperature change rate. It is used to simultaneously characterize the heat load caused by continuous high-temperature operation of the compressor and the thermal shock caused by rapid changes in discharge temperature. The time-of-use electricity price penalty factor is the ratio of the current real-time electricity price to the preset average electricity price.

[0047] The three-dimensional comprehensive performance index is determined according to the following formula:

[0048] In the formula, It is a three-dimensional comprehensive efficiency index; Instantaneous energy efficiency ratio; This is the preset maximum energy efficiency ratio; The thermal stress coefficient of the compressor; This is a time-of-use pricing penalty factor; , and These are energy efficiency weights, thermal stress weights, and electricity price weights, respectively, and they satisfy the following:

[0049] Each weight is adaptively adjusted based on the waste heat margin of the heat source, the compressor exhaust temperature, and the real-time electricity price.

[0050] When the heat source has sufficient waste heat, i.e., the evaporator outlet temperature is more than 5°C above the antifreeze lower limit, the system prioritizes improving energy efficiency and can adjust the energy efficiency weighting. Set to 0.7 to adjust the thermal stress weight. Set it to 0.2 to adjust the electricity price weight. Set it to 0.1.

[0051] When the compressor discharge temperature exceeds 100℃ for 10 consecutive minutes, it indicates that the compressor is under high thermal stress. Therefore, the thermal stress weighting should be adjusted accordingly. Increase the energy efficiency weight to 0.5. The value is reduced to 0.3, and the electricity price weight is adjusted accordingly to keep the sum of the three weights at 1, thereby prioritizing the reduction of compressor thermal stress.

[0052] When the real-time electricity price exceeds the preset peak electricity price threshold, the electricity price weight will be adjusted. The value is increased to 0.4, and the energy efficiency weight and thermal stress weight are reduced accordingly, so that the optimized control prioritizes reducing the compressor input power during periods of high electricity prices.

[0053] Therefore, by integrating instantaneous energy efficiency, equipment health status, and time-of-use electricity price economy into a unified dimensionless index through the three-dimensional comprehensive efficiency index, the problem of excessive energy consumption caused by compressors operating in high thermal stress areas or high electricity price periods due to using only instantaneous energy efficiency ratio as the optimization target is avoided.

[0054] Step S5: Establish a multi-objective constrained optimization model based on the equivalent thermal resistance-heat capacity model, dynamic prediction time domain, and three-dimensional comprehensive performance index; solve the multi-objective constrained optimization model to obtain the optimal control sequence.

[0055] The control variables in the multi-objective constrained optimization model include compressor speed, heat source-side circulating pump frequency, and secondary network circulating pump frequency. When a heat storage tank is configured, the opening degree of the plate heat exchanger bypass valve can also be used as an additional control variable.

[0056] The optimization objectives of the multi-objective constrained optimization model include secondary network water supply temperature tracking error, compressor speed variation penalty, and low three-dimensional comprehensive efficiency index penalty. Among them, the secondary network water supply temperature tracking error is used to ensure stable heating on the user side; the compressor speed variation penalty is used to limit frequent compressor speed increases and decreases, reducing mechanical and thermal shocks; and the low three-dimensional comprehensive efficiency index penalty is used to avoid continuous operation of the system under low energy efficiency, high thermal stress, or high electricity price conditions.

[0057] The corresponding cost function can be expressed as the sum of the weighted terms of the squared tracking error of the water supply temperature in the prediction time domain, the weighted terms of the squared rate of change of the compressor speed, and the penalty term for the low three-dimensional comprehensive efficiency index. The weights of each term can be set according to the user-side heating requirements, equipment protection level, and economic operation requirements.

[0058] During the optimization process, constraints were also set for the upper and lower limits of compressor speed, the upper and lower limits of heat source side circulation pump frequency, the upper and lower limits of secondary network circulation pump frequency, the allowable range of water supply temperature, the allowable range of evaporator outlet temperature, and the allowable range of suction and discharge pressure ratio.

[0059] Furthermore, the compressor minimum frequency limit, evaporator minimum temperature rise protection, and condenser maximum differential pressure change rate are embedded as nonlinear constraints into the multi-objective constraint optimization model. For constraints that may not be strictly satisfied in the short term due to sudden changes in the heat source, relaxation variables are introduced, and corresponding relaxation penalty terms are set in the cost function to ensure that the optimization problem has a feasible solution, while keeping the system as far away as possible from the compressor surge region, the evaporator low temperature danger region, and the condenser differential pressure change region.

[0060] This embodiment employs an improved particle swarm optimization algorithm to solve a multi-objective constrained optimization model. In each control cycle, the current system state estimated by the equivalent thermal resistance-heat capacity model is used as the initial state, and multiple particles are generated within the constraint boundaries of the control variables. Each particle corresponds to a set of control sequences in the dynamic prediction time domain or the control time domain. The control sequences include the compressor speed, heat source-side circulating pump frequency, and secondary network circulating pump frequency at each moment.

[0061] The cost function of the multi-objective constrained optimization model is used as the fitness function of the particles. The velocity and position of the particles are iteratively updated based on the individual optimal position of each particle and the global optimal position of the particle swarm. The improved particle swarm optimization algorithm adopts an inertia weight that adaptively decays with the number of iterations, giving the algorithm strong global search capability in the early stage of iteration and high local optimization accuracy in the later stage of iteration.

[0062] To meet the real-time control requirements of the heating system, the maximum number of iterations of the particle swarm optimization algorithm is set to no more than 30, and an early termination threshold is also set. The iteration is terminated early when the improvement of the optimal value of the fitness function over multiple consecutive iterations is less than the preset threshold.

[0063] To prevent the particle swarm from getting trapped in local optima, a chaotic mapping perturbation is applied to some particles when the global optimum has not been updated during multiple consecutive iterations. The intensity of the chaotic mutation gradually decreases with the number of iterations to balance escaping local optima and maintaining convergence stability in later stages.

[0064] After solving the problem using an improved particle swarm optimization algorithm, the optimal control sequence for the compressor speed, heat source-side circulating pump frequency, and secondary network circulating pump frequency within the future control time domain is obtained. The control system executes only the first control vector in the optimal control sequence, and at the next sampling time, it re-acquires operating data, updates the system state, and solves the optimization problem again, thus forming rolling time domain control.

[0065] Step S6: Execute the first control vector in the optimal control sequence, update the system state based on the deviation between the actual output value and the model prediction value, and perform rolling optimization.

[0066] Because the equivalent thermal resistance-heat capacity model is a simplified model, and there are environmental disturbances, equipment performance degradation and unmodeled dynamics during the operation of the heating system, there may be deviations between the model predictions and the actual output values.

[0067] This embodiment employs an output deviation feedback correction method. Assuming the current measured condenser outlet temperature is the actual secondary network water supply temperature, it is compared with the condenser outlet temperature predicted by the equivalent thermal resistance-heat capacity model to obtain the current output deviation. Under the assumption of constant deviation, this output deviation is considered to remain unchanged within the current dynamic prediction time domain, and the output deviation is used to uniformly correct all subsequent predicted temperatures within the dynamic prediction time domain.

[0068] The corrected predicted temperature is used to recalculate the water supply temperature tracking error and the cost function of the multi-objective constrained optimization model, thereby eliminating the steady-state control deviation caused by model parameter errors.

[0069] In addition to output deviation feedback correction, the state variables of the equivalent thermal resistance-capacity model are updated every 30 minutes using the Kalman filter algorithm. The Kalman filter algorithm determines the relative weights of the state prediction error and the measurement error based on the model's predicted state and the actual measurement data, thereby obtaining the corrected model state. This enables the equivalent thermal resistance-capacity model to adapt to changes in ambient temperature, heat source quality, pipeline thermal inertia, and the slow degradation of equipment performance.

[0070] By combining output deviation feedback correction with Kalman filter state update, model predictive control can achieve unbiased tracking without establishing a high-precision physical model, and improve the accuracy of water supply temperature prediction.

[0071] Step S7: Based on the compressor thermal stress coefficient, evaporator outlet temperature and heat pump suction-discharge pressure ratio, a protection judgment is made, and the operating parameters of the compressor and heat source side circulation pump are adjusted when the preset protection conditions are met.

[0072] This embodiment incorporates independent compressor thermal stress and anti-surge linkage protection logic outside of model predictive control. This protection logic can be set in a programmable logic controller or an independent safety control channel, and its control priority is higher than that of the model predictive control output. When the protection conditions are met, the control commands output by the model predictive control are unconditionally overridden or restricted.

[0073] The first protection channel is used for compressor thermal stress protection. When the compressor thermal stress coefficient exceeds 0.85 and lasts for 5 minutes, the compressor's rate of increase is limited to no more than 0.5 Hz / s, and the compressor frequency is reduced according to a preset frequency reduction cycle. In this embodiment, the compressor frequency is reduced by 1 Hz every 2 minutes until the compressor thermal stress coefficient drops below 0.6. After the thermal stress coefficient returns to a safe range, the forced frequency reduction state is released, and the compressor operating frequency is gradually restored according to the limited rate of increase.

[0074] The second protection channel is used for active surge protection. When the evaporator outlet temperature is below 32°C and the heat pump suction-discharge pressure ratio exceeds 5.5, it is determined that the compressor has entered the preset surge danger zone. The compressor frequency is controlled to decrease linearly at a rate of 0.2 Hz / s, while the frequency of the heat source side circulation pump is controlled to increase at the maximum allowable increase rate to increase the evaporator side flow and improve the evaporator heat exchange conditions.

[0075] The duration of the anti-surge adjustment can be set to 10 minutes. Within 10 minutes, when the evaporator outlet temperature returns to the safe range or the suction-discharge pressure ratio decreases to the preset safe range, the active anti-surge mode will be exited. If the adjustment continues for 10 minutes and the system still fails to leave the danger zone, an alarm will be triggered, and the corresponding compressor will be shut down according to the system safety level.

[0076] Step S8: Perform auxiliary scheduling of the thermal storage tank.

[0077] In this embodiment, the heat storage tank is used as a heat buffer device between the heat pump unit and the user-side heat load, and the operating status of the heat storage tank is incorporated into the model predictive control.

[0078] The change in the equivalent state of charge of the thermal storage tank is determined by the following formula:

[0079] In the formula, This represents the equivalent state of charge of the thermal storage tank. Improve the heat storage tank's charging efficiency; The heat release efficiency of the heat storage tank; The power required to charge the heat storage tank; This refers to the heat release power of the heat storage tank. This is the equivalent heat capacity of the heat storage tank.

[0080] The equivalent state of charge (SPC) of the thermal storage tank can be calculated by integration based on the supply and return water temperatures, the effective volume of the tank, and the specific heat capacity of the water. The cost function of the model predictive control includes a SPC tracking term and an electricity price incentive term, and incorporates the heat release rate of the thermal storage tank into the optimization variables or constraints.

[0081] During periods of low electricity prices and when there is surplus waste heat from the heat source, the control system actively increases the heat pump load, using low-priced electricity to convert low-temperature waste heat into high-grade heat and charge the heat storage tank. In this embodiment, the low electricity price period can be from 11 PM to 7 AM the next day, and the temperature at the top of the heat storage tank is limited to below 83°C during the heat storage process.

[0082] During peak electricity price periods, the heat stored in the thermal storage tank is released first, reducing the real-time output power of the heat pump unit. In this embodiment, the peak electricity price period can be from 7 PM to 9 PM. The heat release rate of the thermal storage tank can be controlled by adjusting the opening of the bypass valve of the plate heat exchanger between the primary and secondary sides, and the bypass valve opening is included as an additional optimization variable in the model predictive control.

[0083] By using thermal storage tanks for auxiliary scheduling, some of the heat load during periods of high electricity prices can be transferred to periods of low electricity prices. At the same time, when the temperature of the heat source suddenly drops, the thermal storage tanks can be used to quickly replenish the heat, thereby reducing the fluctuation of the secondary network water supply temperature.

[0084] To verify the control effect of the method of the present invention, a comparative experiment was conducted for 30 consecutive days in the above-mentioned waste heat recovery heating system of the sintering ring cooler. The control group used traditional proportional-integral-derivative temperature difference control, while the experimental group used the energy efficiency control method of the present invention.

[0085] In a scenario of drastic fluctuations in heat source temperature, the sintering machine underwent a changeover operation between 9:00 and 9:30 AM, during which the heat source temperature dropped from 54°C to 42°C within 3 minutes. Using traditional proportional-integral-derivative (PID) temperature difference control, the secondary network water supply temperature dropped to a minimum of 44.2°C, with a slow recovery rate. Using the method of this invention, the dynamic prediction time domain is automatically shortened during the rapid decrease in heat source temperature. The model predictive control reduces the slope of the condenser outlet water setpoint change in advance and controls the heat storage tank to instantly replenish heat, maintaining the secondary network water supply temperature at a minimum of 47.5°C. The water supply temperature recovery time is reduced by 60% compared to traditional control methods.

[0086] During peak electricity price periods, traditional proportional-integral-derivative (PID) control still operates the heat pump according to a fixed temperature difference, maintaining a high compressor output power. This invention, by increasing the weight of the electricity price penalty factor in the three-dimensional comprehensive efficiency index, actively reduces the compressor load by 15% while simultaneously increasing the heat release rate of the heat storage tank, resulting in a 32% reduction in electricity consumption during peak periods, and ensuring that the indoor temperature of the heated building does not fall below 18°C.

[0087] Regarding the compressor's health status, after a week of continuous operation using a traditional control system, the daily peak discharge temperature of the compressor remained around 108℃, with slight oil carbonization observed. When using the method of this invention, by embedding the compressor's thermal stress coefficient into a three-dimensional comprehensive efficiency index and a multi-objective constraint optimization model, the compressor is actively limited to operate under extreme loads, keeping the peak discharge temperature below 99℃. Throughout the entire test cycle, the compressor's vibration intensity remains within the optimal operating range.

[0088] The results of the continuous tests are shown in the table below:

[0089] As can be seen from the above test results, the present invention, through online identification of the equivalent thermal resistance-heat capacity model, dynamic prediction time-domain adjustment, construction of three-dimensional comprehensive efficiency index, multi-objective constraint optimization, output feedback correction, compressor safety protection and heat storage tank auxiliary scheduling, can maintain stable heating on the user side when the low temperature waste heat source fluctuates significantly, improve the average energy efficiency ratio of the system, reduce the energy consumption of the circulating pump and the operating electricity cost, and reduce the peak temperature of the compressor exhaust.

[0090] The energy efficiency control method of this invention can be embedded in existing programmable logic controllers, distributed control systems, or edge computing gateways. The data acquisition module obtains system operation data from field sensors, heat pump controllers, power metering devices, and electricity price publishing platforms; the model identification module executes a recursive least squares algorithm; the prediction time domain determination module calculates the dynamic prediction time domain; the efficiency index determination module calculates the three-dimensional comprehensive efficiency index; the optimization solution module executes an improved particle swarm optimization algorithm; the feedback correction module completes output deviation correction and Kalman filter state updates; and the protection control module implements compressor thermal stress and anti-surge protection.

[0091] The online model identification and improved particle swarm optimization algorithm has a single control cycle computation time that is shorter than the corresponding control cycle length on an industrial-grade processor, which can meet the requirements of real-time control.

[0092] For low-temperature heat sources with temperatures below 30°C, two-stage series heat pumps or flash tanks can be added to the existing system as auxiliary upgrading devices. The control method of the present invention can still adapt to changes in system structure through online identification.

[0093] For flue gas waste heat recovery scenarios, a corrosion-resistant intermediate heat exchanger can be installed before the evaporator, and the corresponding thermal resistance in the equivalent thermal resistance-heat capacity model can be adjusted to the total thermal resistance including the thermal resistance of the intermediate heat exchanger. The recursive least squares algorithm can re-identify the corresponding model parameters based on the operating data.

[0094] For scenarios involving multiple user areas with different heating demands, the single-node equivalent thermal resistance-capacity model can be extended into a multi-node equivalent thermal resistance-capacity network. Each user area corresponds to at least one equivalent thermal resistance-capacity node. Furthermore, a weighted sum of temperature deviations among user areas is added to the multi-objective constrained optimization model to achieve heat distribution and collaborative optimization among different areas.

[0095] In the system implementation, the energy efficiency control system of the low-temperature waste heat recovery heating system includes a data acquisition module, a model identification module, a prediction time domain determination module, an efficiency index determination module, an optimization solution module, a feedback correction module, and a protection control module.

[0096] The data acquisition module is used to collect parameters on the heat source side, condenser side, compressor operating parameters, environmental parameters, and real-time electricity price, and to perform peak noise removal, low-pass filtering, and missing data repair.

[0097] The model identification module is used to identify the model parameters of the equivalent thermal resistance-heat capacity model online based on system operation data using a recursive least squares algorithm with a forgetting factor.

[0098] The prediction time domain determination module is used to determine the dynamic prediction time domain of the model predictive control based on the rate of change of available waste heat power of the heat source and the prediction error of the heat load.

[0099] The efficiency index determination module is used to determine the three-dimensional comprehensive efficiency index based on the instantaneous energy efficiency ratio, compressor thermal stress coefficient and time-of-use electricity price penalty factor, and adjust the weight of each factor according to the heat source waste heat margin, compressor exhaust temperature and real-time electricity price.

[0100] The optimization solution module is used to establish a multi-objective constrained optimization model based on the equivalent thermal resistance-heat capacity model, dynamic prediction time domain and three-dimensional comprehensive efficiency index, and to use an improved particle swarm algorithm to solve for the optimal control sequence of compressor speed, heat source side circulating pump frequency and secondary network circulating pump frequency.

[0101] The feedback correction module is used to execute the first control vector in the optimal control sequence, correct the subsequent predicted outputs based on the deviation between the actual condenser outlet temperature and the model predicted temperature, and use the Kalman filter algorithm to update the state variables of the equivalent thermal resistance-heat capacity model for rolling optimization.

[0102] The protection control module is used to make protection judgments based on the compressor thermal stress coefficient, evaporator outlet temperature and heat pump suction and discharge pressure ratio. When the preset protection conditions are met, it limits the compressor acceleration rate, reduces the compressor frequency or increases the frequency of the heat source side circulation pump.

[0103] Each module can be implemented by an independent hardware controller, or it can run as a computer program functional module within the same programmable logic controller, distributed control system server, or edge computing gateway. Data transmission between modules occurs via fieldbus, industrial Ethernet, or internal data interfaces, thereby enabling online identification, predictive optimization, feedback correction, and safety protection of the low-temperature waste heat recovery heating system.

[0104] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. An energy efficiency control method for a low-temperature waste heat recovery heating system, characterized in that, Includes the following steps: Collect parameters from the heat source side, condenser side, compressor operating parameters, environmental parameters, and real-time electricity price to obtain system operating data; Based on system operation data, the model parameters of the equivalent thermal resistance-heat capacity model are identified online using the recursive least squares algorithm. The dynamic prediction time domain of the model predictive control is determined based on the rate of change of available waste heat power of the heat source and the heat load prediction error. The three-dimensional comprehensive efficiency index is determined based on the instantaneous energy efficiency ratio, the compressor thermal stress coefficient, and the time-of-use electricity price penalty factor. Based on the equivalent thermal resistance-heat capacity model, dynamic prediction time domain and three-dimensional comprehensive efficiency index, a multi-objective constrained optimization model is established for solution to obtain the optimal control sequence of compressor speed, heat source side circulating pump frequency and secondary network circulating pump frequency; Execute the first control vector in the optimal control sequence, update the system state based on the deviation between the actual output value and the model prediction value, and perform rolling optimization; The protection decision is made based on the compressor thermal stress coefficient, evaporator outlet temperature and heat pump suction-discharge pressure ratio. When the preset protection conditions are met, the operating parameters of the compressor and the heat source side circulation pump are adjusted.

2. The energy efficiency control method for a low-temperature waste heat recovery heating system according to claim 1, characterized in that, The parameters on the heat source side include the evaporator inlet temperature, the evaporator outlet temperature, the heat source flow rate, and the heat source temperature change rate. Condenser-side parameters include condenser inlet temperature, condenser outlet temperature, and secondary network circulation flow rate; Compressor operating parameters include compressor speed, input power, discharge temperature, suction temperature, and suction-discharge pressure ratio; Environmental parameters include outdoor temperature and wind speed; The collected data is processed by removing spike noise, low-pass filtering, and repairing missing data to obtain system operation data.

3. The energy efficiency control method for a low-temperature waste heat recovery heating system according to claim 1, characterized in that, The equivalent thermal resistance-heat capacity model includes a first-order or second-order equivalent thermal resistance-heat capacity link that characterizes the thermal dynamics of the heat source side, heat exchanger, heating network and heated building, respectively. The model parameters are updated using a recursive least squares algorithm with a forgetting factor. When the rate of change of the heat source temperature exceeds 2℃ / min, the forgetting factor is reduced to below 0.

92. When the heat source temperature stabilizes, the forgetting factor is restored to above 0.

98.

4. The energy efficiency control method for a low-temperature waste heat recovery heating system according to claim 1, characterized in that, The dynamic prediction time domain is determined according to the following formula: In the formula, For the dynamic prediction time domain at the current moment, For the time domain of prediction, The rate of change of available waste heat power from the heat source. The nominal waste heat power, For heat load prediction error, For nominal heat load, and For adjustment coefficients, and These are the lower and upper bounds of the prediction time domain, respectively; the lower bound is 15 minutes and the upper bound is 120 minutes.

5. The energy efficiency control method for a low-temperature waste heat recovery heating system according to claim 1, characterized in that, The three-dimensional comprehensive performance index is determined according to the following formula: In the formula, It is a three-dimensional comprehensive efficiency index. Instantaneous energy efficiency ratio, To preset the maximum energy efficiency ratio, The thermal stress coefficient of the compressor, As a time-of-use electricity pricing penalty factor, , and These are energy efficiency weight, thermal stress weight, and electricity price weight, respectively. The weights are adjusted based on the waste heat margin of the heat source, the compressor exhaust temperature, and the real-time electricity price.

6. The energy efficiency control method for a low-temperature waste heat recovery heating system according to claim 1, characterized in that, The optimization objectives of the multi-objective constrained optimization model include water supply temperature tracking error, compressor speed change penalty, and low three-dimensional comprehensive efficiency index penalty; An improved particle swarm optimization algorithm is used to solve the multi-objective constrained optimization model. The improved particle swarm optimization algorithm adopts an inertial weight that decays with the number of iterations, sets an early stopping threshold and a chaotic mutation operator, and applies chaotic mapping perturbation to some particles when the global optimum has not been updated in several consecutive iterations.

7. The energy efficiency control method for a low-temperature waste heat recovery heating system according to claim 1, characterized in that, Based on the deviation between the condenser outlet temperature measured at the current moment and the predicted temperature of the equivalent thermal resistance-heat capacity model, the subsequent predicted temperatures in the dynamic prediction time domain are uniformly corrected; the state variables of the equivalent thermal resistance-heat capacity model are updated using the Kalman filter algorithm according to the preset update cycle, which is 30 minutes.

8. The energy efficiency control method for a low-temperature waste heat recovery heating system according to claim 1, characterized in that, When the compressor thermal stress coefficient exceeds 0.85 and lasts for 5 minutes, the compressor acceleration rate is limited to no more than 0.5 Hz / s, and the compressor frequency is reduced according to the preset frequency reduction cycle until the compressor thermal stress coefficient is reduced to below 0.

6. When the evaporator outlet temperature is below 32℃ and the heat pump suction-discharge pressure ratio exceeds 5.5, the compressor frequency is reduced at a rate of 0.2Hz / s, and the frequency of the heat source side circulation pump is increased until the preset danger zone is removed or an alarm is triggered.

9. The energy efficiency control method for a low-temperature waste heat recovery heating system according to claim 1, characterized in that, It also includes auxiliary scheduling steps for the thermal storage tank, which determine the change in the equivalent state of charge of the thermal storage tank according to the following formula: In the formula, This represents the equivalent state of charge of the thermal storage tank. and These are the heat charging efficiency and the heat releasing efficiency, respectively. and These are the charging power and the discharging power, respectively. The equivalent heat capacity of the thermal storage tank is used; the thermal storage tank is controlled to charge during periods of low electricity price and when the heat source has surplus heat; the thermal storage tank is controlled to release heat during periods of peak electricity price; and the equivalent state of charge and heat release rate of the thermal storage tank are incorporated into the multi-objective constrained optimization model.

10. An energy efficiency control system for a low-temperature waste heat recovery heating system, characterized in that, include: The data acquisition module is used to collect parameters from the heat source side, condenser side, compressor operating parameters, environmental parameters, and real-time electricity prices to obtain system operating data. The model identification module is used to identify the model parameters of the equivalent thermal resistance-heat capacity model online based on system operation data using a recursive least squares algorithm. The prediction time domain determination module is used to determine the dynamic prediction time domain of the model predictive control based on the rate of change of available waste heat power of the heat source and the prediction error of the heat load. The efficiency index determination module is used to determine the three-dimensional comprehensive efficiency index based on the instantaneous energy efficiency ratio, compressor thermal stress coefficient, and time-of-use electricity price penalty factor. The optimization solution module is used to solve a multi-objective constrained optimization model based on the equivalent thermal resistance-heat capacity model, dynamic prediction time domain and three-dimensional comprehensive efficiency index, and obtain the optimal control sequence of compressor speed, heat source side circulating pump frequency and secondary network circulating pump frequency. The feedback correction module is used to execute the first control vector in the optimal control sequence, update the system state based on the deviation between the actual output value and the model prediction value, and perform rolling optimization. The protection control module is used to make protection judgments based on the compressor thermal stress coefficient, evaporator outlet temperature and heat pump suction and discharge pressure ratio, and adjust the operating parameters of the compressor and heat source side circulation pump when the preset protection conditions are met.

Images